Skip to main content

Full text of "Place names, Highlands & islands of Scotland"

See other formats















- \^' 






Dr Alexander Macbain's work on Names of 
Places deals with the Cehic names of pre-Gaehc 
origin which he calls " Pictish " ; with Gaehc 
names, ancl with names of Norse origin which 
have been transmitted through Gaelic. The area 
from which he took his materials was chiefly 
Inverness-shire, Sutherland, and Lewis. His 
views on the language spoken by the Picts are 
given in his paper on " Ptolemy's Geography of 
Scotland" (published separately), in his edition 
of Skene's "Highlanders of Scotland," and in 
several papers contained in this volume, particu- 
larly that on the " Place-Names of Inverness- 
shire." His position is that the Picts spoke 
Early British or a dialect of it, and that the Celtic 
language of early Britain was practically homo- 
geneous from the English Channel to the very 
North. He agrees with Kuno Meyer in holding 
that " no Gael ever set his foot on British soil 
save from a vessel that had put out from Ire- 
larid." Further, assuming that the terms 
Cruthen (which is the Gaelic form of Briton) and 
Pict are co-extensive and mutually convertible, 


he includes under ' ' Picts ' ' the whole of the 
Celtic settlers in Britain prior to the Belgae, thus 
ignoring the facts that the Picts are not heard of 
till about 300 %^, and that all old authorities 
(Gildas, Nennius, Bede, &c.) state that their 
original seat in Scotland was in the far 
North. To him too, as well as to othei 
modern writers, the Cruithne of Ireland are 
"Picts." These assumptions do not, how- 
ever, alfect the linguistic part of Macbain's 
argument, and his views on the language of the 
Picts have been generally accepted. No one 
nowadays would suggest, as Sir John Ehys did 
once, that the Picts spoke a language that was 
non-Aryan, and very few^ w ould hold that Pictish 
was other than Early British. It must be 
admitted at the same time that some of Macbain's 
"Pictish" examples are really Gaelic {e.g. 
Dores, Loch Oich), or at least capable of being 
explained from Gaelic {e.g. Feshie, Mashie, 

By his treatment of Norse names, Macbam 
laid a firm foundation for further investigators to 
build on. He was the first to recognize in 
practice that the changes undergone by these 
names in the mouths of Gaelic speakers are not 
arbitrary, but are capable of classification, and 
that no derivation which ignores the current 
Gaelic pronunciation, or which goes against it, 
can be accepted as certain. 


In dealing with place-names in general, Dr 
Macbain's method was first to make sure of the 
actual pronunciation in Gaelic, and then to com- 
pare the old written forms of the names when 
such are available. He also paid attention to 
the physical characteristics in cases where there 
niight be more than one possible explanation. 
In the case of Idrigill, for instance, I remember 
how he learned first that there is no " gill 
there, and then that there is a knob-like hill at 
the extremity next the sea. This is, of course, 
the only scientific way of treating the subject. 
He was not always right, and in the papers that 
follow it will be seen that he changed his views in 
particular instances; but he was on the right 
lines. Sometimes he was misled by wrong 
information : this is most notable in his paper on 
ISutherland. The only work on Scottish place- 
names comparable to Macbain's, done by a man 
of his generation, is the late Professor Donald 
Mackinnon's series of eighteen articles in the 
Scotsman on the Place-Names and Personal 
Names of Argyll, which has not been reprinted. 

The present volume contains practically all 
that Macbain printed on the subject of Place- 
Names. It may be proper to state that before 
the work came into my hands, the selection and 
sequence of the papers had been already decided, 
and pp. 1-64 printed off. 


In the notes that loUow, 1 have indicated the 
chief points on which, as it seems to me, one 
might venture to differ fi'om Dr Macbain's 


I . Cataobh, GaUaobh, being certainly datives, 
are correctly spelled Cataibh, Gallaibh. 

6. Ptolemy's tribes : In his paper on Ptolemy's 
Geography of Scotland, Macbam places 
the Caereni " in Western Sutherland up 
to near the Naver " ; " the Cornavii occu- 
pied Caithness, the horn or corn of 
Scotland " ; " neighbours to the Cornavii 
southwards were the Lugi, occupying 
Easter Sutherland. Around Loch Shin 
were the Smertae, and Easter Ross was 
' occupied, up to the Varar estuary, by 
the Decantae." The name of tiie 
Smertae was discovered (by myself) to 
survive in Carn Smeart (also, sometimes, 
Carn Smeartach), the name of a hill in 
the ridge between Strathcarron and Kyle- 
side. Smertae is a participial form, fiom 
the root siner, smear; compare M. Ir. 
Travedum, Travedrum : read Tarvedum, 

8 . Creed river — A' Ghriota : correctly Abhainn 


Pittentrail : in Gaelic Baile an Traill, 
Thrall's Stead, which makes it post- 
Norse, for train is a loan from Norse 

Aberscross : in Gaelic abarscaig ; Norse ; no 
connection with aber, a confluence. 

10. Eogart : in Gaelic Raghart, i.e. Ro-ghort, 

Ea-ghart, Big-field. 
xlssynt : assendi will not do, for initial a of 
Norse ass, ridge, is long. There is 
another Assynt in Ross-shire . 

11. Skinaskink : a ghost-name fouud on maps 

for the real Sionasgaig. 

Clyne : as Mr CM. Robertson has pointed 
out, Clin, which is the Gaelic form, is an 
oblique case of claon, sloping, a slope. 

Dornoch : place of fist-stones, i.e., rounded 
pebbles or dornagan: the locative case is 
found in Dornaigh, Dornie, in Lochalsh, 
and elsewhere— r. Place-Names of Ross 
and Cromarty. 

Kildonan : Kelduninach c. 1230; in view of 
other early spellings and of the modern 
Gaelic, Cill Donnain, there is no doubt 
that " St Donnan's Kirk " is the original, 
and that Kelduninach (presumably inter- 
preted by Macbain as Kil-domnach) is 
either an error or refers to another place. 

Lairg : in Gaelic Luirg, dative-locative of 
lorg, shank; compare Lurgyndaspok, 



1390, "the Bishop's Leg" (Ant. of 

Aberdeen and Banff); Magh-luirg, Moy- 

lurg, in Connacht. 
Reay : in Rob Donn Mioghiadh, genitive 

Mioghraidh (rhyming with inntinn, cinnt- 

each, gniomh sin) ; in Strathy now Meagh- 

rath ; cannot therefore be from magh ; the 

second part is rath; compare Dii(n)rath, 

Embo : may be Eyvind's Stead; Elvind is 

probably a misprint. 
Creich : in Gaehc Craoich, possibly locative 

of craohhach, tree-place; not from cnoc/i. 
li). Ashore: in Rob Donn, and now, Aisir (a 

long); anglicised on maps Old-shore. 
Hysbackie, in Gaelic Hei(ll)sbacaidh; the 

phonetics clearly indicate an original II 

(or possibly nn); compare Heisker. 
Coldbackie, in Gaelic Callbacaidh. 
14. Migdale : in Gaelic Migein, not Norse, 

wholly at least, and to be compared with 

Migvie, Miggovie, Miggernie, etc. 
Keoldale : in Gaelic Cealldail; the palatal c 

is decisive against kaldr, cold. 
Duible, in Gaelic Daigheabul; the hrst 

syllable is sounded like aoi short. 
Leirable, in Gaelic Lireabol; not from leir, 

loam ; it may be from Norse liri, a tern, 

also a man's nickname. 



Eldrable, in Gaelic Eilldreabol, which can- 
not be from the source suggested in the 

Mudale, in uaehc Modhadul. 

Halladale, in Gaehc Healadul. 

15. Navidale : in Gaehc Nei(mh)eadail, which 
may be a hybrid from 7ieimhidh, a sanc- 
tuary, holy place, and Norse dah, a dale. 

Conamheall : properly Conmheall, either 
High Lump or Hound Lump ; con may be 
the compositional form of cu, hound, or 
it may represent Early Celtic cimos, high. 
As Conmheall is the highest part of Ben 
More in Assynt, it probably represents 
Early Celtic Cunomello- . 

iriimisdale, in Gaelic Eumasdal; the u is 
almost ao short. 

Iti. Meall Rinidh, in Gaelic Meall Eoidhinigh 
or Eeidhinigh (possibly Eoithinidh, 
Eeithinidh) ; Loch More is in Gaelic Loch 
an Eei'inidh; there is also Allt an Eei'- 
inidh; the ei is close. My informant 
connected these names with reidhneach, 
reithneach (in the glossary to Eob Donn, 
1829, reidhne), "bo sheasg," a yeld cow. 
Reisgill : there is another Eeisgiil in Suther- 
land, which is in Gaelic Eidhisgii; I am 
not sure of its position. 



k>migol, ill Gaelic Smidl)i*;ii ; not from 
smiiga; it appears lo represent Norse 
" smidbju-gil," Smithy-gill. 

Fresgill, in Gaelic Freisgii, perhaps froui 
Norse fress, a tom-cat. 

Siiisgil, in Ciaelic Sithisgil or Sidhisgil ; 
Norse " seydhir," from which Macbain 
takes the name, means a lire-pit, cook- 
mgfire, roasting fire. The derivation is 
somewhat doubtful . 

Ben Loyal : in Gaelic Beinn Laghail, Norse 
laga-fjaU or larja-rdJlr, Law-fell or Law- 

TralagiU : Thrall's Gill, not Troll's Gill. 

Baligil : the a is long, therefore Norse 
Bale-gulley, Flame-gulley. 

Melness, in Gaelic Mealanais; cf. Meala- 
bhaig, Bent-grass Bay. 

Shinness : more likely Gaelic sean-innis^ 
old-haugh . 
17. Conesaid, in Gaelic Caonasaid; the phon 
etics point to a Norse '" kein-" ; compare 
" Thorgeir Keingr " (genitive Keings) 
of Landnamabok, where "Keingr" is 
explained as " uncus," hooked. 

Falside, in Gaehc Feallasaid, correctly ex- 
plained in text. 

Melvich, in Gaelic Mealbhaich, Place of 
Seabent; not from Norse vik. 



Golval, in Gaelic Golbhal, where o, being 
short, cannot represent an of Norse, 
which would give o long in Gaehc. 

Musal, in Gaelic Musal; the explanation is 
probably correct. 

Marrel, in Gaelic Maraill, Sealield. 

Rossal : the grass here causes congestion 
and inflammation in cattle, but not m the 
case of horses. 

1."^. Hielam, in Rob Donn Huilleum; may he 
Hound -holm. 

Scourie; in Gaelic Sgohhairidh, probably 
from Norse skogr, a wood, " Shaw- 

Bighouse : in Gaelic Biogas, genitive Biog- 
ais; an alternative explanation is Norse 
hygg, barley : Barley-house. 

Olave : in Altas, a fancy name given to a 

32. "A well called Dobur Artbranani ': read "a 
stream," etc. 

" Losing himself in a dense wood " : read 
" entering a dense wood." 

Clar Sgithe : in poetry clar is common in 
the sense of " surface, district," e.g. 
Clar Chormaic, Clar Conghail, Clar Cobh- 
thaigh, etc. (bardic names for Ireland), 
Clar Monaidh (North Britain); Clar 
Mumhan (Munster); Clar Rois (Ross); 
Clar Fionnghall (the Western Isles, 



Hebrides), etc., etc. '' Clar Sgith," 
therefore, is in no way influenced by 
Norse skidh. The term occurs often in 
unpubhshed bardic poetry, regularly in 
the form of Cldr Sgith or Sgi. 

33. Dun Sgathaich : read Dim Sgathaich; so in 
Skye now ; the Dean of Lismore has 
genitive ^'caf/ic/?a, rhyming with cjnrdha. 
The poem in the Dean's Book has also the 
dative Scdthaigh rhyming with d' fhdg- 
ais; both forms occur m Early Irish. 

35. Swordale : in Gaelic Snardail; Norse Saur- 
dalr would become Sordail in Gaelic. It 
is " Sward-dale." 

Sleat : the derivation from sleitr is sup- 
ported by the spelling SUite in an unpub- 
lished poem by Cathal MacMhuinch. 

Bracadale : in Gaelic Bracadail (Brachda- 
dail); therefore not from hrekka, a slope. 

36.Eaasay: in Gaelic Ratharsa; MacVurich 
Raarsaigh (genitive); v. p. 169. 

Trodday : in Gaelic Trondaidh, evidently 
connected with Trdiidairnis, but 
" Thrond's Isle " ought to be Trondar- 

Ascrib. Isles : initial a is long, which nega- 
tives the explanation. 

Wiay : in Gaelic Fuiaidh, which does not 
consist with the explanation. 



37. Broadford : a translation of an i-Atli 
Leathan, the Broad Ford; not Norse. 

Oskaig : oss-skiki, Stream-mouth Strip. 

Loch Eishort : Gaehc, as got by me, Loch 
Ai(ll)seort, does not consist with the ex- 

Scavaig : in Gaelic Sgathabhaig; hardly 
from Norse skogr, wood, shaw 

Osdail : initial o short negatives the deriva- 

39. Garry : in Gaelic gearraidh, Norse gerdhi, 
a fenced field, garth. 

46. Kilmallie : the suggested derivation from 

Amhalghaidh is impossible; Mailidh is 
most probably connected with mdi, a 
prince; found also in Con-e Mhaileagain 
(Place-Names of Eoss and Cromarty) and 
in Dail Mhailidh, Dalmally . 
An Linne Sheilich : read ' ' iVn Linne 
Sheileach," which is from setle, locally 
said to mean here ' ' brackish water ' ' ; 
compare seile, saliva. 

47. Loch Leven : I have heard it called " Loch 

Liobhunn," but the true form appears to 
be " Loch Leamhain " or "Leamhna," 
from leamhan, elm. The river Leven m 
Lennox is certainly Leamhan; so in the 
poem by Muireach Ua Dalaigh, " Saer do 
lennan, a Leamhain." Glen Lyon is 
Gleann Lio{h}i)unn. 



4.S. Glen Loy : in Gaelic Gleann Laoigh; Lao^h, 
calf, is the name of the river. A well at 
Tara was called Loig-les, " vitulus civi- 
tatum," Calf of the Courts. 
Callart : in Gaelic CaUaird, Hazel Point. 

49. Dun Dearduil : onlv two forts of this name 
are known to me, one at Inverfarigaig on 
Loch Ness, the other this one, in Glen 
Bothuntin : the local pronunciation in Gaelic- 
is Both-thionntainyi . 

^)''>. Loch-ais' : the lingering, dragging sound 
that indicates the loss of // is attached to 
the i, not to the a. 

•">*'. Ulhava : if Ulfr were a person's name, it 
ought to be rather Ulfs-ey; in Gaelic, 
Ulbhsa; but the s of the Norse genitive 
seems to be dropped sometimes in the 
Gaelicized forms. 

57. Avernish : afar, bulky, seems to be used 

only of qualities and actions, not of 

things like nesses. 
Loch Calavie : in Gaelic Loch Cailbhidh, 

from calbh, a plant-stalk, etc. 
Strathasgaig : in Gaelic Srath-asgaig ; Norse 

d-skiki, water-strip. 

r>". Conchra : more probably "Dog-fold"; 
compare Ir. con-chro, a wolf -trap; K. 
Meyer's Contribb. 



61. Strathie : as it has the singular form of the 
article, " abhainn an t-Srathaidh," the 
form must be diminutive. 
Palascaig : in Gaelic FeaUasgaig, Norse 
Fjalla-skiki, Hill-strip. Palasgaig was 
formed on the assumption that / was 
aspirated p. 
Monar : the phonetics are decisive against 
the reference to monmhur; the Gaelic is 
Monar, from root of mon-adh, hill, moun- 

63. Loch Hourn : the couplet quoted from the 
Dean's Book is in Deibhidhe metre, and 
should be read — 

Leigid deireadh do mhuirne 
eadar Seile is Subhairne. 

They make an end of jollity between 
Sheil and Subhairne. 

Tohn MacCodrum has — 

cheann Loch-Uthairn nam fuar 

Gu bun na stuaighe a Morrair, 

which I have printed in Bardachd 
Ghaidhlig "Loch Shubhairn," following 
the Dean. At the head of the loch there 
are Coire Shuhh and Loch Coire Shuhh, 
from suhh, small fruit, berry, raspberry, 
etc. ; and Subhairn is therefore Suhh- 
hhearn, Berry Gap. Similarly A' Mhor- 



hhairn, Morverii, is from mor, the com- 
positional form of muir, sea, and hearn, 
" the Sea Gap," with reference probably 
to the deep indentation by Loch Sunart. 

68. Sainea : the equation with Shuna is impos- 

sible phonetically, for Sainea would be- 
come Saoine in modern Gaelic, while 
Shuna is now Siuna. 
Maleos : read Malaios. 

69. Colonsay : in Gaelic Colhhasa; in the 

Dean's Book, " jholfissay " ; in the 
so-called " Eed Book of Clanranald," 
" Colbhannsaigh " ; with Dean Monro, 
" Colvansay " ; on record " Coluynsay,'* 
etc. — plainly Norse, " Kolbein's Isle." 
Adamnan's Colosus is now Coll. 

Terra Ethica : there is no reason to doubt 
the equation with Tii'ee, but the reference 
to Old Irish ith , oenitive etho, corn, is 
doubtful, in view of the fact that Tiree 
appears in very old Irish poetry as Ti'r 
lath, which indicates that e of " Ethica " 
is long. 
78. Harris: another suggestion is Norse 
heradh, a district, but the fact that e of 
heradh is long while that of Na h-Earradh 
is short is fatal to this idea. 

Hirt : i\ p. 177; perhaps the more probable, 
if more orosaic, connection with Old Irish 
irt, death, is the extremely dangerous 



character of the rock-bound coast of 
Hirt. The term recurs in an Duihhirt- 
each, " the Black Deadly One," the 
name of the sea-rock west of Colonsay on 
which a lighthouse now stands. 

74. Coilsay : read " Gilsay." 

75. Fuidhaidh : /•. note on p. 36 above. 
Benbecula : the Gaelic form is pioperly 

Beinn na bJiFaoghJa. 

Heisker : in Gaelic Hei{U)s(jeir: old spell- 
ings have / regularly; the name is Norse 
hellu-sher, flat skerry, contracted into 

Hasker : in Gaehc Haisgeir, haf-sker, deep- 
sea skerry. 

76. Trodday : the derivation from trodh, pas- 

ture, is impossible, for Norse f//^ would be 
in (-aelic here r/h. See note on p. 36 
above. The correct explanation of 
Trotternish appears on p. 166. 
Eaasay : r. p. 169; Hraunsey is not pos- 
sible . 

77. Kerrera : in Gaelic pronunciation there is 

still distinct trace of hit — Cear(bli)ara; 
kjarr, copse, is therefore insufficient lo 
explain the phonetic facts. 

78. Colonsay : see note on p. 68 above. 
Hersey : it is difficult to see how this could 

have been a Norse attempt at pronouncino- 



Bute : in modern Gaelic rather Bod ; Baile 
Bhoid is Eothesay; there is Old Irish hot, 
fire, but the connection is not obvious. 
79. Snizort : v. p. 34. 

81. Creed : this is all wrong; see p. 8 and note 

thereon above. 

82. Diebek : this is the same as Diobaig, 

" Deep Bay," in Eoss-shire. 

83. Leurbost : cannot come from leir, which 

would yield laor in Gaelic. 
Garbost : cannot come from geir, which 
would yield gaoi' in Gaelic. 

84. Habost : the derivation proposed by Captain 

Thomas' is impossible, for II would not 
disappear here. 

85. Haugr : becomes Hogh and Togh in Gaelic. 

88. Bragar : in Gaelic Bragar. 

89. Orfris-ey : read Orfiris-ey. 

90. Eodel : in an Adv. Lib. MS. Roghadal 

(poem of 1705). 

92. Taransay :Taran was a Pictish personal 

name; see Index to Skene's Chronicles of 
the Picts and Scots. 

93. Eesort : in GaeHc Reusort; Capt. Thomas'.^ 

explanation is impossible phonetically . 

95. Coinn-mheall : see note on p. 15 above. 

96. Gardhr : throughout this article, read 

gerdhi, a fenced field, garth. The Gaehc 
is gearraidh. 



Rusigarry : in Gaelic Ruisigearraidh , which 
makes a derivation from hris improbable. 
97. Tralagill : in Gaelic Tralaigil, Thrall's 

Crisigill : the explanation "Cross-gill" is 
phonetically unlikely. I do not know the 
Gaelic pronunciation. 
100. " Joyce is wrong " : Norse muli would not 
become rnaol, maoil in Gaelic; Joyce is 
right so far as the Mull of Kintyre is con- 
cerned, and probably also as to the Muli 
of Galloway. The MiiUs of Orkney, etc., 
represent mxili doubtless. 

103. Callernish : in Gaelic Calanis; " Caia " is 

found at the beginning of several Lewis 
names, and probably is the personal 
name Kali. Mr Kenneth Mackenzie of 
Shader, Barvas, believed the form CaUer- 
nish to be wrong ; it is mdicated, however, 
by Martin's " Classerniss." In any case, 
kjalar-nes does not suit the fact that c of 
Calanis is not palatal, as it would be if it 
were Norse kj. 

104. Barvas : the derivation is certainly wrong 

as to the first part, and very doubtful as 

to the second part. 
107. Linshader : in Gaehc Liseadair, with nasal 

i; Flax- stead. 
109. Vatisker : the first part has no connection 

with vddha; probably for vains-sker, 



water-skerry, i.e., skerry covered at high 
Hasker : Deep-sea Skerry, haf-sker. 

112. Clach an Truiseil : derivation unsatisfactoiy 

phonetically; a more likely one will ap 
pear in Mr Kenneth Mackenzie's forth- 
coming book on the Flace-Names of 

113. Teangue : rather from Gaehc teanga. 
121. Inverness: in the Dean's Book, Inverness 

is " v'nvir nissa," i.e. Inbhir Nise (the 
metre requires a dissyllable); in 17th 
century bardic poetry (poetry by profes- 
sional trained bards) it is Inbhir Nis, 
without the genitive inflection, as now. 
Clachnaharry : Clach na h-Aithrigh(e), 
Stone of Repentance, suits the phonetics 
exactly; cf. Clach a' Pheanais, in Colon- 
Tomnahurich : in literature iuhhrach is used 
in the sense of a " barge, goodly vessel" ; 
also " a pretty girl." 
126. Bona : well known in Gaelic as am Banath, 
the White ford, and named so, according 
to the local seanchaidhean, from white 
stones in it. Similar formations are com- 
mon, e.g., an Damhath, Dava, the Ox 
Ford; an Garhhath, Garva, the Rough 
Ford; am Bannath, Bonar, the Bottom 
Ford, etc. Macbain's suspicioi] was 


INTR0DUCT10>. XXlll. 

perhaps due to the fact that there is no 
ford at Bona now, but that is due to the 
raising of the level of Loch Ness when the 
Caledonian Canal was made. 

Dores : see note to p. 157. 

Moy : The Mackintosh is " Tigliearna na 

Nairn : in Gaelic Narunn ( PNarrunn) ; kSiath 
Narunn and Srath Naruinn, Strathnairn; 
Inbhir Narunn and Inbhir Naruinn, Nau'i], 
town. The name belongs to the small 
but important class of river names -hat 
end in -ami, -unn, e.g., Comhann, Coe; 
Liobhunn, Lyon ; Carrann, Carron. Mac- 
bain regarded these as representmg the 
Early Celtic ending -ona: they may, how- 
ever, be names of river divinities in {he 
genitive case, from an old nomniaiive 
ending in -u; compare Domnu, gen. 
Domnann; Manau, gen. Manann. 

128. Croy is in Gaelic Crothaigh. 

Partick is in Gaelic Pearraig (for Pearth- 

Blairour : the fact that the confluence of 
the Blair-our burn with Spean is Inver- 
our indicates that Blair-our represents 
Blar-dhobhair, Moor of the Water, i.e. 

129. Pelier : there are other streams of this 

name in addition to those mentioned. In 



" Inbhir-feo'arain," / is of course really 
ph. The final -an (open a) is not in- 
flected; i.e., it is not -ain. This lack of 
inflection in the case of the final -an (from 
-agnos) is common in the early language, 
and in some parts of Scotland is still 
regular in saints' names, e.g. Ciaran, etc. 

147. Don : I have not heard Dian. 

151. "A medigeval MS." : the Book of Lemster 
(circa 1150 a.d.), 371, b. 37 : " Donnan 
Ega — Ega nomen fontis i n-Aldasain i 
Cataib i tuasciurt Alban ' ' ; Ega is the 
name of a well in Aldasain among the 
Cats (i.e. in Sutherland or Caithness) m 
the north of Alba. An interlinear gloss 
on Aldasain says correctly, " .i. carrac 
etir Gall-Gedelu 7 Cend-tiri i n-a camair 
immuich " ; "(Aldasain is) a rock be- 
tween Galloway and Kintyre facing them 
out (in the sea) " ; it is Ailsa Craig in fact. 
With " Ega nomen fontis " compare 
" enga, aqua super petram, .i. fons," 
' ' enga, water over a rock, that is, a foun- 
tain" (Book of Armagh), where " enga " 
might be written, in Greek style, 

Morvern, which has o short, cannot con- 
tain w,6r ; the first S3^11able is the composi- 
tional form of miiir, sea ; Morvern means 



" Sea-gap," just as Subhairne means 
' ' Berry-gap. ' ' See note on p. 63 above. 
Portree : there is another Fort-righ in Km- 
tyre, a very old name. The Skye name 
may be much older than James V. 
152. Morbhearnaibh : see note to p. 63. 

157. Duthil : anglicized from Gaelic Daoghal or 

Daodhal ; similarly Culduthel, near Inver- 
ness, is Cuil-daoghail, -daodhail; this puts 
" tuathail " out of the question. 

Dores, in Gaelic Duras . there is no phonetic 
difficulty m equating this with duhhros, 
duhhras, which in fact occurs as Durrus 
in Cork (Joyce). Terminal -as here is 
dull, whereas the so-called ' ' Pictish ' ' 
ending -ais has open a. Dorus, locally 
darus, a door, is the Gaelic name which 
is anglicized Dares, in the parish of Dores. 
The local rhyme beginning ' ' Mile o 
Dhuras gu Darus," " a mile from Dores 
to Dares," is very well known. 

Loch Hourn : see note to p. 63 above. 

158. Dalarossie, Dulergusy : the ending -ie or -y 

represents the old genitive ending of 
Fergus, an u-stem with genitive Ferguso, 
164. Loch Hourn : see note to p. 63. 

159. Ben Loyal : see note to p. 16 above. Mac- 

bain's leidh-fjall would yield laodhcd in 



172. (Jreagarry or Oreagorry is Creag Ghoraidh, 

Godfrey's rock. 
176. Elrick : in the Book of Deer " elerc, 

which comes by metathesis from Old 

Irish erelc, an ambush. 
178. Lochmaddy : in Gaehc Loch nam Madadh, 

Loch of the Dogs ; the '' ' dogs ' ' are 

three rocks hi the bay, called " na 

Boisdale : in Gaehc Baoghasdail, Baegi's 

Gieniinnan : in Gaelic Gleann Fhionghiiin, 

Fingon's Glen; from Fingon comes Mac- 

Fhionghuin, anglicized Mackimion. 

180. Loch Arkaig : in Gaelic Loch xiirceig. 
Glenquoich : ' ' cuach ' ' may have refer- 
ence to pot-holes in the river; the fact 
that " cuaich " is singular is no objection, 
being quite in keeping with Gaehc usage. 

Loch Oich : it is sniiply Loch Obhaich, for- 
merly Loch Abhaich, Loch of Abhach, 
i.e. stream-place, from ahh, river, 
whence Awe. That the stream novv- 
known as the Oich river was once Abha 
is proved by the fact that a stretch of 
ground above it is called Fachdar Abha, 
the Overland of Awe. 

181. Phoineas in Kiltarlity has no waterfall neai 

it; it 1-=^ "fo-innis," suh-meadov, small 
meadow or haugh. 



Ooylum : from cumg-leam, ' gorge-leap '' ; 

cong, a deep narrow gorge in a stream 

(or even between rocks m the sea), is with 

us cuiyig. 

Rothiemoon : " Eat a' Monie ' ' ; we should 

have expected " na mona." 
Geldie, etc. : it is to be feared that here we 
have Old Irish gelda, geldai, bright; e.g., 
" Aedan in grian geldai," in Feillire 
Oengusa ; ' ' Aedan the brilliant sun ' ' ; and 
elsewhere passim. 

188. Scaniport : the stress being on the first part, 
the meaning is rather " Cleft-ferry," i.e. 
ferry near the cleft. 

319. Ach-gourish represents " gobhair-innis," 
Goat-haugh, Goat-mead. Compare Coin- 
innis, Hound-mead; Daimh-innis, Ox- 

327. This is a review of the first edition of the 
work; the second edition (1903) benefited 
by the criticism. 

331. abh, O.Ir. ab, means stream, river; it is 
feminine; genitiye " na habae, nahaba," 
in Earlv Middle Irish, 
"an, water" : so Kuno Meyer in Contnbb., 
with reference to Stokes' ' ' Metrical 

334. " Dal-uar, with the accent on the Dal " : 
the stress would be on " uar," the quah- 
fying term, not on " Dal," the generic 



term. The fatal objection to connecting 
names like Bal-four, Pit-four, with Gaelic 
" fuar," cold, is that when these names 
are preserved in Gaelic pronunciation, the 
sound of " four " is not in the least like 
that of "fuar'." 

343. " The Gael did not visit the Epidii for at 
least forty years later." Macbain means 
that Cairbre Eiada's settlement took 
place about forty years after the time 
when Ptolemy wrote. 
But Gael from Ireland may have 
visited ' ' Kintyre much earlier ; the 
regulus ' ' who visited Agricola in 
Scotland about 84 a.d. is not likely to 
have been the first to come across. 

346. Maol-rubha : the declension of the name 
(e.g., mac Maile-Rubha in Cain Adam- 
nain) negatives mdl, prince. Mael-rubha 
is exactly equivalent to Maei-ruis or 
Mael-rois, from ros, cape or wood; in 
such names ma el has lost its primitive 
meaning of " cropped man, shaveling," 
whence " slave, devotee " ; and is prac- 
tically equivalent to gille. 
Glen Finnan, see note, p. xxvi. 

348. Ardnamurchan : the fact that the name is 
stressed on the penultimate syllable puts 
" Heishts of the sea of Coll '" out of the 



question; the grammar, too, is impos- 

350. Colonsay : see note, p. xviii. 

352. Nant is simply neannta, nettle; Abhainn 
Neannta, Nettle river; Coille Neannta, 
Nettle M^ood, etc. 

354. Glen Brander (Branter) : I have not heard 
the name pronounced, but the Pass of 
Brander is Cumhang a' Bhrannraidh, 
from h7'annradh, an obstruction; Irish 
brannradh, a trap or snare; stocks, pil- 
Seil, in Book of Leinster 24b, Soil; more 
likely pre-Norse; O.N. seil means a 
string, which is not satisfactory as an 
island name. 

C O N T K N T S. 


Introduction . . . . . .v. 

Sutherland : its Early History and Names . . 2 

{Highland Neiv.s, 26th March, 2nd April, 9th 
April, 16th April, 1898). 

Place Names of Skye ..... 32 

(Highland Newt:, 11th December, 1897). 

Place Names of Lochaber ..... 42 
(Highland News, 12th February, 1898). 

Place Names of Lochalsh ..... 53 
(Highland News, 2nd February, 1901). 

Place Names of the Hebrides .... 67 
(Northern Chronicle, 16th, 23rd, and 30th Novem- 
ber, and 2nd December, 1892). 

Place Names of Inverness and Vicinity . .121 

(Northern Chronicle, 12th July, 1893). 

Place Names of Inverness-shire . . . .133 

(Transactions of the Gaelic Society of Inverness, 
Vol. XXV.; read 13th February, 1902. Also 
Northern Chronicle. 1899— 15th March, 22nd 
March, and 29th March). 

Badenoch : its History, Clans, and Place Names . 189 
(Transactions of the Gaelic Society of Inverness, 
Vol. XVI.; read 5th March, 1890). 

Annat 283 

(Northern Chronicle, 20th May, 1903). 

Gaelic " Airigh," Shieling, in Norse Place Names 289 
(Northern Chronicle, 29th Jiily, 1903). 

Glenshiel 296 

(Northern Chronicle, 2nd September, 1903). 



Tonuiaburich 299 

(Northern Chronicle, 2nd September, 1903). 

Place Names of Ross and Croiririrty (Review) . 303 
(Celtic Review, July, 1904). 

The Place Names of Elginshire (Review) . . 313 

(nighUind News). 

Place Names of Scotland (Review) . . . 327 

(Inverness Courier, 1st Miixch, 1892). 

The Place Names of Argyll (Review) . . . .*^?,9 

(Highland News, 28tli April and 5th M:iy, 1906). 

Index ........ 361 




THE name Sutherland was applied by the Norse to the 
portion of their Caithness south of the Ord, 
stretching to the Oikel river, now the southern 
boundary of the modern county. The water-shed 
in the middle of the county divided this Sudhr-land 
from Assynt, Durness, and Strathnaver ; and the northern 
district of Strathnaver and its neighbouring valleys were 
known to the Norsemen as the Dales of Caithness. This 
restricted meaning of the name remained in historic force 
till 1601, when the Earl of Sutherland got the modern 
county, all save Assynt, raised mto a separate Sheriffship, 
apart from Inverness, in the Sheriffship of which it had 
been till then included. The Earl of Sutherland's lands 
also were till then mostly confined to the district here 
indicated as early Sutherland. Through the fall of the 
Roman Church, which practically possessed north-west 
Sutherland, and through the turbulence of the native clans 
— the Mackays especially — the Earl of Sutherland in 1601 
was either actual holder or legal superior of the present 


county, with the exception of Assynt. In 1631 Assynt 
was also joined to the rest of Sutherland, and the present 
county was constituted, thanks to the efforts of the 
indefatigable Sir Robert Gordon. 

The Gaelic name for Sutherland is Cataobh, and Brae- 
Chat is Sir Robert Gordon's designation for the upper 
regions of Lairg — the Barony of Gruids and the other 
inland and upland districts on the eastern water-shed, with 
Dirie-Chat, or the Desert of Cat, further north. The 
Norse called both Sutherland and Caithness by the name 
of Caithness or Katanes ; but when greater accuracy was 
desired, modern Caithness was called Ness, that is, the 
Nose of the province of Cat, while the district south of 
the Ord w^as called Sudhr-land. This distinction remained 
after the Norse power was overthrown, and w^e hear of 
" Catanesia cis et ultra Montem " — Caithness on this and 
beyond the Mound — the MONTEM being the Ord of Caith- 
ness. In an important epitome of the geography of 
Scotland, written in 1165, and inspired by Andrew, the 
first Bishop of Caithness, we read : — " Septima enim pars 
est Cathanesia citra montem et ultra montem, quod mons 
Mound dividit Cathanesia per medium." The division 
made by the Ord of Caithness suggested the usual deriva- 
tion given for the Gaelic names of Sutherland and Caith- 
ness — the names Cataobh and Gallaobh. They were 
explained by Shaw, the historian of Moray, as being for 
Cat-taobh and Gall-thaobh, the CAT-side and the Gall- 
side of the Ord ; Gall he explained as stranger or foreigner, 
and this is correct, and Cat he derived from Gaelic CAD, 
high, which is a non-existent word, or from St Cattan's 
name. As a matter of fact the two names are the dative 
or accusative plural of the Gaelic nouns Cat and Gall. The 
name of a people was in old Gaelic times used for the 
name of their country, in the plural number, and generally 
in the dative or accusative plural. The same thing occurred 
in Latin and in Anglo-Saxon. The name Wales means 


" Welshmen ;" it is a plural of like force with Gall or 
Gallaobh, foreigners, or " Lowlanders " now. Caithness 
in Gaelic is Gallaobh, and means " strangers," " among 
strangers," "in the land of strangers," that is, AN 

But what is Cataobh ? Gaelic " cat " means a cat, as 
in English. Various interpretations have been offered. 
Careless investigators have correlated the name with the 
Chatti of Germany, mentioned in the first century of our 
era, and they have even asserted that Ptolemy places a 
Catti tribe in Sutherland. The tribe meant is the 
Decantae, which has been misread into Cantae, which even 
then is unlike Catti, and still more unlike the Chatti, or 
rather Hatti, now developed into the province of Hesse. 
The name Cat, Cait, or Cataobh is old ; it is manifestly 
antecedent to the Norse, who made use of the tribal names 
they found on the mainland of Scotland — the Picts for 
Pettland, as in Pentland Firth ; and so Cat and Cata was 
the name adopted for the province, which was divided 
into the Ness of Cat and the Souther-land. Katanes or 
Caithness latterly usurped the older name of Kata or Katar, 
which, as a matter of fact, does not exist in historical Norse 
literature ; it is only inferred. The Gaelic records of 
mediaeval times make Cat a son of Cruithne, the eponymus 
of the Cruithnig or Picts ; he was one of the seven sons of 
Cruithne, who divided Scotland between them, and a verse (X-m^rru^f 
is recorded which is attributed to St Columba, which says : 

Seven children of Cruithne 
Divided Alba into seven divisions — 
Cait, Ce, Cirig, a warlike clan, 
Fib, Fidach, Fotla, Fortrenn. 

The four provinces of Cataobh, Fife, Athole (or old Ath- 
Fhodhla), and Fortrenn are clearly indicated ; the other 
three names are difficult to fix. Mediaeval Irish works 


refer to the northern province of Scotland as Crich Chat 
and I CATAIB, " in (the land of) Cats," or m the Province 
of Cats. There is every indication that the name ante- 
ceded the Norse ; and, further, the word CAT, cat, possessed 
by most modern languages, is probably Celtic, meaning 
the " wild cat." The name appears in Gaulish as a per- 
sonal name — Cattos and Catta ; and there can be little 
doubt that the Gaelic name of Sutherland comes from an 
old Celtic tribe — the Catti — named so after the " wild cat." 
Such animal names were quite common as tribal names, 
and it is supposed that the Caereni (Assynt, probably) 
mentioned by Ptolemy as possessing western Sutherland 
were so named from CAORA, sheep. 

Ptolemy, the geographer, who flourished about 120 of 
our era, has left half-a-dozen interesting names attachable 
to Sutherland and Caithness. The tribal names are the 
Caereni, already mentioned ; north of them, in the north- 
west corner of the county, occupying Durness, were the 
Cornavi, a name which might mean the people of the 
CORN or horn, as we have in Cornwall. The Lugi (or 
Lougi) inhabited eastern Catanesia ; and south of them 
were the Decantae, with the Smertae to the west, possibly 
of both. The latter name is from the root SMER, MER, 
mind, memory, which also appears elsewhere, especially 
in Ro-smerta, the Gaulish Minerva. These five tribes, 
with the Carnonacae, in the Loch-carron and adjacent 
districts, filled the whole of Scotland north of the Beauly 
Firth. Ptolemy mentions the capes, rivers, and estuaries. 
Curiously, Cape Wrath is missed; but the river Naver is 
given plainly enough in his " Nabarus flumen," the root of 
the name being supposed to be NAV, flow, swim, as in our 
borrowed English words, NAVAL, NAVY. The Cape of 
Travedum or Travedrum may be regarded as one of the 
horns of Thurso Bay ; the word means " bull head," from 
TARVOS, now TARBH, bull. That this is the locality meant 
seems to be proved by the name Thurso, older Thorsa, for 


Thjorsa, bull water, a name which appears also in Iceland. 
The name Orkas is also given to this cape, and Orkney is 
the Norse garb of the Celtic name Orcades or Whale-isles. 
Ptolemy gives also two other cape names further east: 
Duncansbay Head is Cape Virvedrum, and Noss Head is 
Verubium, which last possibly means " Spit-head." The 
Helmsdale river, Gaelic Ilidh, is rendered with fair exact- 
ness by Ptolemy's Ila flumen. With this name may be 
compared the other Isla of Perthshire, and Dr Stokes has 
proposed the root IL, EIL, move rapidly, as the root of the 
name. Further south Ptolemy places his Alta Ripa or 
High Bank, which is supposed to be the Ord of Caithness 

One or two misreadings of Ptolemy, besides supposi- 
tious names which the perfervid imaginations of anti- 
quarians have conjured up, are responsible for some bad 
history and bad etymologising. The case of the name 
Chatti or Hatti has already been mentioned ; but a bad 
reading of the name Lugi has been adopted, namely, Logi, 
which is made to explain the parish name of Loth. The 
Lougi may have been so named from the Celtic or Gadelic 
sun-god, whose name was Luga, possibly meaning the 
" bright and charming one," perhaps allied to the Norse 
Loki, the god of tricks and evil deeds. A name Abona 
has somehow " growed " in this connection, and it is made 
to explain Bonar. 

The first Celtic inhabitants of Sutherland were the 
Picts ; it was from them that the names recorded by 
Ptolemy came. They spoke a language like the Welsh, 
where P often answers to Gaelic C. The great test-word 
in place-names is PET, whose Gaelic equivalent is CUID. 
It signifies a farm or " town," the same as Gaelic BAILE, 
which, in fact, replaces it. Half-a-dozen names with the 
prefix PIT or PET meet us in Sutherland proper, for there 
are none now in the north and west of the county. There 


is, first, Pitfour; this is a common name in Pictland 
(occurring twice in Sutherland — in Rogart and near Lairg), 
and means, in all probability, " Pasture-town," allied ta 
Welsh PAWR, pasture. Pitmean, in the old Barony of 
Skelbo, is also common, and possibly means " Mid-town," 
allied to Gaelic MEADHON, middle. It appears as Pait- 
mayne in 1525 and Petmayne in 15G2. Pettakarsie and 
Pitfour are mentioned together in 15G6. Pitgrudie shows 
a terminal part which seems to appear in Gruids and the 
Grudie river in Durness. It has been explained as from 
the Norse GRJOT, gravel ; and as a river name it is classical 
in the Cumberland form of Greta. In Lewis we have it in 
the Creed river — A' Ghriota. But it can scarcely be the 
same word in Pitgrudy ; such a hybrid is almost impossible 
in the circumstances. The Pictish language by the time 
of the Norse conquest of Sutherland was practically dead ; 
a new combination with PET under Norse auspices is 
scarcely to be imagined. The old form of the name, in 
1222-45, was Pethgrudi, which is unlike what we should 
expect from Norse GRJOT at that time, or indeed later. 
A form like GREED or GREOD is demanded by Norse- 
Gaehc phonetics. Compare the river name Fleet from 
Norse FLJOTR, fleet, flood, and the clan name Macleod, 
which comes from Norse LJOTR, ugly. The Welsh GRUT, 
of similar meaning, has been suggested, but the history of 
the Welsh word itself requires clearing up. Pittentrail (in 
1566 Pittentrail) has been explained as Pet-an-traigh — 
the town by the shore ; but this does not account for the 
ending of the word. 

The only other assured Pictish names are Abirscor and 
Oykel. In regard to Abirscor, there being Easter and 
Wester Abirscor, the word is generally plural — Aberscors, 
now Aberscross. The natives pronounce it Aberscaig. 
The etymology of the name still awaits elucidating. In 
1518 it is spelt Abbirsco, in 1544 Abirscor. That it is 
Pictish is proved by the prefix ABER, instead of the Gaelic 


equivalent INVER, meaning " confluence." The Oykel 
river is probably so named from its banks : the name in 
Pictish means " high," and is the same as appears in the 
Ochil Hills and m Ochiltree ("High-town" or Uxello- 
trebos). The Norse sagas speak of Ekkjalbakki or the 
Oykel bank, and this fact also lends strength to the view 
that the river got its name from the high banks some- 

The Norsemen commenced their raids shortly before 
800. At first they did not think of settling in the land. 
About 830 they began to establish a kingdom in Ireland, 
and they had evidently meanwhile subdued the Orkneys 
and Hebrides and colonised them, intending them as so 
many stepping-stones in their raids in Ireland and the west 
of England. The conquest of Sutherland and Caithness 
is recorded as having taken place about 880 under Thors- 
tein the Red and Sigurd of Orkney. The Norse had pos- 
session of the province of Cat for over three hundred 
years. It was not till 1196 that King William finally estab- 
lished the authority of the Scottish Crown north of the 
Oikel ; for it was only a nominal suzeranity that existed 
previous to that, and Earls like Thorfin (1014-1064) were 
quite independent of the King of Scotland ; indeed, the 
latter ruled as a rival — a friendly rival— to Macbeth, pos- 
sessing all Scotland north of the Beauly Firth, if, indeed, 
his power did not extend to Inverness. The name Ding- 
wall (" Parliament-place ") shows that they had estab- 
ilshed a centre of political authority there. The complete- 
ness of the conquest of Sutherland is shown by the great 
number of Norse place-names that still exist therein. In 
Cosmo Innes' map of Sutherland, attached to his " Origines 
Parochiales," where all the " public " names of the county 
appear as they were in the 16th and 17th centuries, the 
proportion of Norse names in Sutherland proper — Dor- 
noch, etc. — is one in every three as against Gaelic, while 
in the " Dales " district — Tongue, etc. — the proportion is 


reversed, and Gaelic forms only a third of the names as 
against Norse. The conquest and occupation of Suther- 
land proper were slower and less complete than what took 
place on the northern coast. As we come through Easter 
Ross the Norse names fall away rapidly, and end alto- 
gether in the Beauly valley with Tarradale and Eskadale. 

In northern Sutherland we meet with as many Norse 
place-names nearly as in Lewis. The general Gaelic 
names STAC, CLEIT, GEODHA and SGEIR, meaning respec- 
tively a precipitous hill, sea cliff, bay, and " skerry," are 
from the Norse ; and these are common names along the 
coast of Sutherland from Assynt to Reay. Beinn Stack 
in Eddrachilles is one of the highest mountains in the 
county. Of the thirteen or fourteen parish names in 
Sutherland three are certainly Norse — Durness, which 
means " Deer's Ness or point," spelt about 1230 as Dyrnes ; 
Golspie, spelt in 1330 Goldespy, which is a compound of 
the Norse BAER or BYR, a village, English BYE, as in 
Whitby, and whose first part is possibly GULL, gold, though 
usually explained as Gils-by or " Ravine-village ;" and 
Tongue, which is from Norse TUNGA, a tongue (of land). 
Rogart is possibly Norse ; its earliest spelling (1230) is 
Rothe-gorthe, then Rogert in 1542, and Roart in 1562, 
which is practically its local name still. It seems to be 
for Raudhar-garth or Rauth's garth or farm, the RAUDH 
signifying a person as well as " red." It is explained 
usually as ROTH-GART, " circle garth " in Welsh, as being 
of Pictish origin. If GARTH forms the final part of the 
name, it is infallibly Norse ; for DAL and GARTH or GARRY 
(from GERDI) final are of Norse origin and use. Assynt 
has also been claimed as Norse, explained as A SYNT, " seen 
from afar," with reference to its conspicuous mountains. 
The earhest spelling is Assend, and for this we might 
suggest the derivation ASSENDI, the Norse for "Ridge- 
end." Its termination may be compared to that of the 
Caithness parish name of Skmnet, the older Scynend, the 


Skinid of Tongue, which may be compared with the 
Icelandic Skinnastadr or " Skinstead ;" and to these we 
may add the name of Loch Skinashink in Assynt. 

The other parish names are Celtic or Gaelic. Clyne is 
the Clun of 1230 from G. CLUAIN, a meadow. Creich is 
Crech in 1230, and doubtless means the " boundary " 
parish— G. Crioch ; Dornoch, which in 1230 is Durnach, 
seems of Pictish origin, pointmg to a Celtic Durnacon, the 
stem DURNO appearing both in England and on the Con- 
tinent in Celtic place-names and meaning " stronghold " 
doubtless allied to Gaelic DORN, fist; Eddrachilles means 
the place " between the two kyles ;" Farr, a name also 
appearing in Strathnairn, is possibly a compound from G. 
FOR, over, above, and means " upper land ;" Kildonan was 
originally Kil-domnach or " Lord's Kirk," corrupted into 
"St Donnan's Kirk ;" Lairg, in 1230 Larg, means a "hillside 
or moor," G. LEARG ; Loth is from G. LOTH, mud, now 
obsolete ; Reay, which is partly of Caithness and partly 
of Sutherland, is in G. Rath or Magh Rath, " Plain of 
the Fort " (there is also Ben Rath), reminding us of the 
Ulster Maghrath or Moyra, famous in story. The earliest 
form of the Gaelic name appears in M'Vurich, who calls 
the first Lord Reay "Morbhair Meghrath," Mormaer of 

In rapidly reviewing the Norse names of Sutherland, 
I will first commence with personal names which enter 
into place-names. Persons' names often give names 
to farms, especially with the word BOL, farm. Thus 
Arnaboll of Durness, which appears in the sixteenth 
century as Ardeboll, Arnboll, Ardnaboll, and Arnobill, 
means Ami's stead, rather than " eagle or erne stead ;" 
Embo, which in the early ITth century is Enbo, and in 
1610 Eyndboll, means Elvind's stead, even though the 
modern pronunciation is Ereboll, a manifest and easily 
explained corruption ; Skibo is the Scitheboll of 1230, 


which suggests the name Skithi, the word SKIDH meaning 
otherwise a log, tablet ; Torboll of Dornoch, which appears 
as Torroboll in Lairg, appears as Thoreboll or Thureboll 
in the 13th century, and is so named from Thori, a fav- 
ourite Norse name, derived from the god named Thor. 
Another form of this name is Thorir, genitive Thoris, 
which appears in Torrisdale of Tongue. Names in pre- 
fixed Thor or Tor are very common all over the Norse- 
occupied portion of the Highlands and Isles. Unapool of 
Assynt receives its name from Uni; it means Uni's BOL. 
Allied to BOL is the word BOLSTADR, farm-stead; it 
becomes BISTER or BUSTER in Caithness and Orkney, and 
BOST in the Isles. Ulbster of Kildonan no doubt was 
Ulli's stead, a favourite nnme, which also appears in Ulla- 
pool and Ulladale elsewhere ; while Scrabster (in Tongue), 
which in the Orkney Saga appears as Ska-ra-bolstadr, 
means Skari's stead rather than " sea-mew stead," which 
it may also mean. With DALR, a dale, personal names are 
rare ; yet we have Helmsdale in the Sagas as Hjalmundal, 
which means Hjalmund's dale. Ospisdale, in Creich, is 
from Ospis, which must be the genitive, degraded consi- 
derably, of Ospak or Uspak, another favourite Scoto- 
Norse name. Ullipsdale, in Kildonan, is doubtless 
" Wolf's dale," after Gaelic phonetics had hardened the F 
of Ulfs (genitive of Ulfr) into a P before the S. Trantle, 
in Farr, which appears in 1527 as Trountal and in 1626 as 
Trontaill, stands no doubt for a Norse Throndar-dalr, or 
" Thrond's dale ;" and doubtless the same name accounts 
for the Dronside or Thrond's SETR (seat) of Tongue. 
Dal-Harald, in Farr, is a Gaelic compound, and, as conjec- 
tured, commemorates the defeat in 1196 of Earl Harold, 
son of Maddad (Gaelic MADADH — Hound) of Athole, by 
King William, when the King was helped by the famous 
Manx King Reginald, son of Godred, who undertook the 
government of Cataobh for a season. Lochan Hacoin, in 
Tongue, is named after some unknown or unrecognised 


Haco or other, just as Kyleakin of Skye celebrates Haco 
of Largs (1263). The name Grimr has left many place- 
names in the Isles ; two places in Sutherland get named 
after some hero of this designation — Ben Griam, Sir R. 
Gordon's Bingrime, and, below the ben, Griamacharry, 
or " Grim's Garth " — that is, Griama-ghardhaidh in 
the best old Gaehc phonetics. Not far away is Ben 
Armin, the ben of the ARMANN, w^hich in Norse means 
steward or controller, and in Gaelic, which has borrowed 
it, " a hero." Cyderhall is a fancy form for Sidera, which 
in 1230 appears as Sywardhoch, in 12T5 as Sytheraw ; it is 
no doubt justly regarded as standing for Sigurd's HAUGR 
or "howe," where the first Sigurd of Orkney may have 
been buried after his fight with and death by Malbrigd 
Bucktooth, whose venomous tooth had killed him. Ashore 
or Oldshores in Assynt was formerly Astlair (1559) and 
Aslar (1551), and the late Captain Thomas regarded this 
as a corruption of Asleifar-vik, Asleif's Bay, which is 
mentioned in 1263 as one of Hacon's ports of call. 

The most of the Norse names will now be classified 
under their commonest, significant parts, such as bakki, 
bol, dalr, and vik. 

Bakki, a bank. We have Backies above Golspie, "the 
Banks ;" Coldbackie in Tongue, which is either Cold Bank 
or Charcoal Bank, probably the former; Hysbackie, also 
in Tongue, for Hus-bakki, "House-bank;" and the Saga 
Ekkjals-bakki or Oikel Bank, where Oikel itself seems to 
be the Pictish UCHEL, high, possibly applied to the river 

BOL, a farm. Arnaboll, Embo, Skibo, Torboll, Torro- 
bol, and Unabol have been explained above. In Dornoch 
parish we have Skelbo, the older Skelbol and Skelbotil, 
which means "shell town (bol or botl) ;" in Durness, Erri- 
bol, "Beach-town," from EYRR, beach, and Loch Crossphuil 
from Krossa-bol ,"cross-town," a name well known in the 


Isles ; in Kildonan, Duible, "Mud town/' from DY, mud, and 
Leirable from LEIR, loam, meaning much the same as 
Duible — old forms of both are Doypull (1527), Duiboll 
(IGIO), Lyriboll, Lereboll (1563-156G), to which compare 
Lerwick, "mud bay ;" Borrobale is Borg or Burgh-bol, 
"fort-town," Borryboll (1563); Eldrable, older Eltriboll 
(1610) and Altreboll (1566), which cannot be from ELDR, 
gen. ELDS, beacon, as usually explained, must be equated 
with the Caithness Alterwall, the Alterwell of 1455, which 
points to a Norse Altara-voUr or "Altar-field ;" hence we 
may infer Eldrable to be for "Altar-ton." Gailval in 
1566 Galezboll, is possibly Galli's town: in Lairg we have 
Colaboll, which may mean " Coal (charcoal) town," or 
" Cold town," or even " Kol's town," the person Kol : in 
Tongue we have Kirkiboll, Icelandic Kirkjubol, " Church- 
town," and Ribigill, which in 1530 appears as Regeboll, 
and may thus mean " Lady's town " (RYGR, lady). 

Dalr, a dale. In Creich there is Swordale, the 
Swerdel and Swerisdale of 1275, meaning " Swarddale ;" 
Spinningdale, in 1464 Spanigidill, and in 1553 Spanzedell, 
possibly " spangle-dale," from Norse SPONG, G. SPANGAR ; 
Migdale, the Miggeweth of 1275, seems from MYKI, dung. 
In Dornoch there is Astle, which has undergone many 
transformations. Askesdale and Haskesdale (1222-75), 
Assastel (1360), Askadaile (1472), Assiedale (1610), which 
is the Icelandic Eskidalr or " Ash-dale." In Durness 
there is Keoldale, in 1559 Kauldale, the Icelandic Kaldi- 
dalr, " Cold-dale ;" Strath-undale, Strathwradell of 1530, 
the dale of the URUS or auroch. In Farr are many dales 
— Armadale, Armidill (1499), " Arm or bay dale ;" Mudale, 
Mowdaill (1570), Mowadale (1601), possibly from MODA 
MODR, muddy river or snow-banks, which seemingly is 
the root idea of Moydart also — "Mudfjord;" Halladale, 
Helgadall in 1222, means " Hallow or Holy dale," though 
the name may be a personal one, Helgi ; Langdale is 
exactly " Longdale." In Kildonan we found Helmsdale 


and Ullipsdale, already discussed ; there are also Navidale 
(Navadaill, 1566) and Rimisdale (Rimbisdell, 1630), the 
former being explained as from NAEFR, birch (compare 
Icelandic Naefrholt, " Birch-holt," and the latter from 
RYMR, roaring, " Dale of the roaring stream." In Lairg 
we have Sletdale, " Evendale," and Osdale (Feith Osdale), 
" East-dale." 

Ey, island. Oldney off Assynt is possibly from 
ALDIN, fruit ; the Channel Island Alderney has been com- 
pared in name. Soyea is Saudhar-ey or " Sheep-isle," a 
common name in the Hebrides ; Chrona is possibly 
T-hraun-ey, the same as Rona, " Rocky-isle." Off 
Eddrachilles are Calva or " Calf -isle," a common name 
also, and Handa, " Sand-isle." Boursa, near Strathy 
Point, is apparently BURS-EY, " Bower-isle." In 1386 
Ferchard Leche, or the Physician, gets from Robert III. 
the islands from " Rowestorenastynghe to Rowearmedale " 
(Rudha-Stor-an-Assaint to Rudha-Armadail), which are 
named Jura (" Deer isle,' possibly Oldaney), Calva (" Calf- 
isle,") Sanda (Handa, " Sand-isle,") Elangawne, Elanwill- 
ighe, Elanerone, Elanehoga, Elanequothra, Elangelye, 
and Elaneneyfe. In 1570 some of these are Handa, 
Choarie (Quothra in 1551), Gyld (Rabbit Isles?), Rone 
(" Seal Isles ") and Colme, while Howga, now Hoan, also 
appears (Haga and Houga in 1601, 1613). The latter 
means the " howe " or " burial " isle ; and itself and Isle- 
Colm or Neave, " Holy-isle," were ancient burial-places 
" to keep the bodies safe from the mainland wolves !" 

FjORDR FJARDAR, a ford or sea-loch. Laxford 
(G. Luiseard) and INCHARD are both on the coast of 
Eddrachilles ; the former means '" Salmon-loch " and the 
latter probably " Meadowf jord," from ENGI, a mead. 

FjALL, hill, fell. This suffix seems to have been 
replaced in Sutherland by BEINN and MEALL of the Gaelic. 
Suilven, from SULA, pillar; Conamheall, from Konna- 


fjall "Lady's fell;" and Far-mheall, from FAER, sheep. 
Ben Arkle must be from Arkfell, from its summit being 
" ark-like ;" Aleall Horn is simply the Norse HORN, which 
is common for hills and capes ; Meall Rinidh may be for 
Hreinn-fjall, " Reindeer fell," for the Norse found rein- 
deer, it is said, in Sutherland. Beinn Loyal has the ter- 
minal FJALL, but the prefix is obscure; compare the 
Icelandic Laufafell or " Leafy-hill." In South Uist is the 
similarly named Ben Layaval. 

Gil, a ravnie. In Assynt we have Tralagill, usually 
explained as Troll's gill, but Thrall's gill is also an Ice- 
landic word, and suits here as well ; Urigill, ravine of the 
URUS or auroch ; Gisgill, the " gushing gill," allied to 
GEYSIR, hot spring. The Reisgill of Eddrachilles is pos- 
sibly from HRIS, brushwood. Farr has Apigill, " Ape-gill," 
which recalls the Icelandic Apavatn, where API may have 
been a person's nick-name; Baligill, gill of the grassy- 
slope (BALI) ; Smigel, gill of the narrow cleft (SMUGA) ; and 
Redigill, possibly Retta-gil, the " gill of the sheep pen or 
adjusting pen." Fresgill, in Durness, is explained as the 
" noisy gill " (FRAES, noise). Suisgil, in Kildonan, is in 
1527 Seyisgill and Suisgill in 1545, which may be com- 
pared to the Icelandic Seythisfjordr, " Seethe-fjord." Mr 
Mackay refers it to SUS, roaring. 

GjA, a rift, geo, G. GEODHA, borrowed. The Gaelic 
form of this word is very common on the northern shores 

of Sutherland ; its Norse use is found in Sango-more and 
Sango-beg, " Sand-bay," in Durness, and Lamigo or 
^' Lamb's bay," in Tongue. 

Nes, a ness, cape. Melness, in Tongue, means 
" Bent-grass-ness " (MELR) ; Unes, in Golspie, " Yew-ness," 
from YR, yew, the Owenes of 1275 ; and Shinness, which 
in 1630 is Chinenes, " Ness of Shin or Loch Shin," called 
in Gaelic Ard-na-sinnis. 


Setr, a seat, farm, sheiling. It often appears as 
SIDE in Sutherland : in Tongue, Conesaid, Konnasetr, 
" Lady's-ton," the older Kinsett (1570) and Kenny side 
(1601) ; Falside, "Hill or fjall seat;" in Creich, Linside, 
Linsett in 1541 and Leynside in 1552, possibly " flax- 
seat " (lin) ; in Golspie, Clayside, possibly " Cliff-seat " 
(KLEIF) ; Bosset and Bowsett (Creich and Farr), " Dwell- 
ing-seat " (BUSETR), while in Reay Sandside, the older 
Sandset, means " Sand-seat." 

ViK, bay. In Assynt are Melvich (" Bent-grass ") and 
Kirkaig (^Kirkjavik, " Kirk-bay ") ; m Farr, Melvich ; in 
Durness, Cearbhaig, or Kerwick, Karfavi, " Galley bay." 
Port Chahgaig, in Eddrachilles, is Cellach or Kjallak's 

Vollr, g. VALLR, a field. Carrol, in Clyne, Carrell 
in 1610, is Kjarr-vollr or Copse-field. Rossal, in Rosehall, 
is Hross-vollr, " Horse-field," Rosswell in 1553 ; Langwell 
and Dal-Langal are both from Langi-vollr or " Longr 
field ;" Sletell, in Tongue, is " Even-field ;" Golvall, in Farr, 
the older Gauldwell (1559), may be Galh's field or Gaular- 
vollr, " field of the sounding stream," Norse Gaular-dale. 
Musal, in Durness, is Moswell in 1560, that is " Mossfield ;" 
.while Majrel, in Kildonan, is explained as " Sea-field." 

Other names that do not often recur and do not 
come under " heads " are these : — Stoer m Assynt, 
which is the very common Norse prefix STOR, big, 
Stor-ass, " Big-ridge," Stor-isandr, " Stour-sand," etc. ; 
Brora (Bruray 1601, Brora sixteenth century), which 
is the Icelandic Bruara, " Bridgewater " exactly ; Uppat, 
Uphald in 1528, is from UPP, up; Kyle-strome and 
Ben-Strome, admissable hybrids of Gaelic and Norse, 
come from STRAUMR, stream, ocean current; Smoo, Cave 
of Smoo, in Durness, from SMUGA, a rift or narrow cleft 
to creep through, SMJUGA, to creep; Rispond, from HRIS, 


copse ; Cape Wrath or Am Parph is from the Norse name 
HVARF, turning point; Hope, ben and loch, from HOP, a 
bay, as in Oban, Ob, etc. ; Hielam is a compound of 
HOLMR, a holm or island, but the old forms are puzzling — 
Unlem (1542), Handlemet (1551), Hunleam (IGOl) ; 
Sandwood in Eddrachilles is for Sandvatn, " Sand-water " 
— it is Sandwat in 1559; Scourie, place of sheds or 
shiels ( ?), from SKURR, a shed ; Borgie in Farr is from 
BURG, a fort, the Borve or Borr of the Isles ; Port Skerra 
and Skerray are from SKER, a sharp rock, whence "skerry" 
and Gaehc SGEIR ; Forsinard, etc., the " twa Fursyis " 
(1527\ Forseyis (1626), are from FORS, a waterfall; 
Swordly is probably " Sward-lea ;" Skullomie may be 
Skolla hvamur, " fox's slope ;" and the following may be 
Norse : — Gearnsary, Grodsary, Modsory, Pronsy, Maikle, 
Sciberscross, Olave, Shigra, Skericha, Syre, Kirtomy, Big- 
house and Garty. Grumbeg and Grumbmore (also Grubeg 
and Grubmor) have been etymologised by Rev. A. Gunn 
as from Druim-beg and Druim-mor, a phonetic impossi- 
bility. The old forms explain their origin : in 1570 we 
have Grubmor and Grubeg, but in 1551 it is Gnowb 
" Mekle and litil," which is the common Icelandic place- 
name word GNUPR, a peak, " a knob." Bighouse is pro- 
bably Bygd HUS, " Dwelling-house." 


Before passing on to the Clan and Family Names of 
Sutherland, I have, firstly, to acknowledge my deep 
indebtedness, in my study of the Norse names of Suther- 
land, to Mr John Mackay of Hereford, whose excellent 


series of papers in the Inverness Gaelic Society's Trans- 
actions on " Sutherland Place-Names " made my task com- 
paratively easy — in fact, made it a matter of judicial, if 
not judicious, selection. 


The oldest family name in Sutherland is that of 
Moray or Murray. The noble family of Sutherland hailed 
originally from Moray ; Freskin, the ancestor of the Earls 
of Sutherland and the great families of Morays of Both- 
well and Tullibardine, whence the Duke of Athole and 
the Earl of Dunmore, held the lands of Duffus, in Moray, 
about 1150. His son, called Hugh Freskin, got Suther- 
land — that is, Sutherland proper — from King William, no 
doubt at the time of his conquest of Catanesia in 1196 ; 
and William, his son, was created Earl of Sutherland about 
1235, much about the same time as Magnus, son of Gille- 
bride, Earl of Angus, was made Earl of Caithness proper, 
Freskin's son William was ancestor of the De Moravia 
family — the Morays — famed in the 13th and 14th cen- 
turies, the best known of them being Sir Andrew Moray 
of Bothwell, the Scottish patriot. 

It cannot be proved that Hugo Freskin was called 
" of Moray," nor was his son so called, for he calls himself 
" Dominus de Suthyrlandia, filius Hugonis Freskyn ;" but 
his friends were Moravians. Gilbert, Bishop of Caithness, 
1222-1245, was " de Moravia," acquiring the lands of 
Skelbo from Hugh Freskin, and latterly granting them to 
his brother, Richard de Moravia of Culbyn, in whose 
family they remained for two centuries. Next to Suther- 
land the name Moray is the commonest met with in old 
documents dealing with Dornoch and the adjacent 
parishes. The Province of Moray is called in early Gaelic 


Mureb, and in Norse Morhaefi, which points to a Pictish 
Mcrapia, roots MOR (sea) and AP (water), meaning " coast- 

As a name Sutherland is naturally very common in 
Sutherland proper. Its origin is simple : it arises from the 
title " de Sutherland," or " of Sutherland." In its older 
form it appears as " Nicholas of Sutherland of Duffus," for 
example. Scions of the noble house only had the name, 
just as " de Isles " or " Isles " did duty for the surname 
of the early IMacdonald Chiefs — Alexander Isles of Glen- 
garry and Marion de Ilys, sister of Alexander, Earl of 
Ross (1439). The name Sutherland is, therefore, on a 
different footing from any other county-named surnames, 
such as Nairne and Fyfe. The tenants of Cupar-Abbey 
lands were sometimes from the neighbouring " kingdom,'^ 
and such are called, for lack of other surnames, Henry of 
Fife and James of Fife, as the case may be, that is, James, 
from Fife, " James the Fifer," which latterly settles into 
James Fife. 

The Gordons became Earls of Sutherland in the 16th 
century on the failure of the Moray family in the male 
line, Adam Gordon, second son of the Earl of Huntly, 
marrying the heiress Elizabeth, and their son Alexander 
being infeft in the Earldom in 152T. The Gordons there- 
after became fairly numerous in the county. The name 
is derived from the lands of Gordon, in Berwick, " de 
Gordon " being its original form as a designation. One 
of the most noted Sutherland men of this name was Sir 
Robert Gordon of Gordonston, tutor of Sutherland from 
1615 to 1630, being uncle to the young Earl. He wrote 
that valuable work, "The Genealogy of the Earls of Suth- 
erland," which is our most important guide for the early 
history of the northern Highlands. 



Sir Robert Gordon gives as the principal surnames 
in Sutherland proper (leaving out Strathnaver, Durness, 
Eddrachilles, and Assynt, which last was joined to the 
county by Sir Robert's efforts m 1631) in his day the 
following: — Gordon, Sutherland, Moray, Gray, Clan-Guin 
(the Gunns), Seil-Thomas, Seil-Wohan, and Seil-Phaill. I 
have already discussed the first three clan names. The 
Grays had their chiet holding at Skibo, which was pos- 
sessed by them in the 16th and 17th centuries. The first 
of them came North from Forfarshire about 1456, being 
the son of Lord Gray of Fowlis, who had to fly for killing 
the constable of Dundee. 

The Clan Gunn were of Caithness origin. The name 
is Norse, the common one of Gunni, from GUNNR, war 
(allied to Eng. GUN). Their ancestor was Gun, crowner 
of Caithness, about the middle of the 15th centur}' (1450). 
He was a man of great power in his day, and a descent 
from the King of Denmark was claimed for him — a son 
of that King called Gunni having settled, ages before 
Sir Robert Gordon wrote the story, in Caithness. The 
Crowner's daughter was mother of Donald Gallach of 
Sleat, slain in 1506, ancestor of Lord Macdonald. The 
Crowner was treacherously slain by the Keiths, along with 
many of his clan. His son James escaped to Sutherland, 
and the chief centre of the clan in Sutherland came to 
be Killernan in Strath-ully. William Mac James Mac 
Crowner Gunn distinguished himself against the Mackays 
in 1517, at the battle of Torran-dow. During the 16th 
century the history of Clan Gunn was a very chequered 
one ; the clan had branches in Caithness, Strathnaver and 
Sutherland proper, and they could not please the rival 
heads of the houses of Sutherland, Sinclair and Mackay. 



They generally sided with the Earl of Sutherland. In 
1585 the earls agreed to destroy the Gunns, and for three 
years the plucky clan was harassed by the earls, until 
another disagreement intervened to save the Gunns. John 
Gunn Robson was leader of the Caithness Gunns in 
IGIC), and no doubt the northern Robsons may be traced 
to this branch of the clan. Colonel Sir Wilham Gunn, 
his son, rendered himself famous in the Continental wars, 
becoming by 1G48 a Baron of the Roman Empire. 

The Seil-Thomas and Seil-Phaill, mentioned by Sir 
Robert Gordon, were cadets of the Mackay clan. There 
is some doubt as to the Seil-Thomas, for Sir Robert, in 
his attempt to wean them from the Mackay alliance, de- 
clared them to be from Ardmeanagh, in Ross. But the 
Seil-Phaill or McPhails were really Mackays, descendants 
of Paul, son of Neil Mac Neil Mackay, the latter Neil 
being a brother of Angus Dubh, the great Mackay Chief 
of 1411. This Neil Mac Neil was the first Mackay to get 
land from the Crown under charter; King James, in 
1430, gave him lands in Creich and Gairloch. The Pol- 
sons, or Paulsons, are supposed to be the same as the 
McPhails, sons of Paul Mackay, but this is not the case. 
The real Poisons go back as such to the 15th century, and 
were settled before 1430 m Creichmore in Creich, where 
John Poilsone is mentioned in 1545, and Thomas Poilsone 
in 15GT. They are regarded by Cosmo Innes as des- 
cended from Paul MacTire, who got these lands in 1365. 
Paul MacTire, or " Paul the Wolf," was a famous man in 
his day. One or two ecclesiastics appear with the name 
of Poison in the 16th century. The name Paul was a 
favourite one among the Norse, and hence its popularity 
in Cathanesia. 



Seil-Wohan, or Seil-Wogan, as Sir Robert Gordon 
has the name, were Siol-Mhathain, or the Sept of Mathe- 
sons. The name Mathan simply means " bear," and was 
extremely common in Ireland, especially in Norse times, 
when Thor's Wolf and Thor's Bear formed the model for 
warlike names. The Mathesons or Mac Mahons were 
located on the north side of Loch Shin, their Chief being 
tacksman of Shiness. They are traditionally regarded as 
descended from the Lochalsh Mathesons, one of whom, 
Donald, son of Alexander, fled for his country's good to 
Caithness or Cataobh. It was his grandson, "John Mac- 
ean-Mac-Konald-wain " whom Sir Robert Gordon in 1616 
induced to assume the chiefship of the sept and to 
separate his name and kin from the Seil-Thomas, who 
were on the side of their kinsmen, the Mackays, and who 
were bringing the Mathesons over to the Mackay side. 
In this way Sir Robert, as he says, weakened the power 
of Seil-Thomas. The first mentioned in the county is 
William Matheson in 1512, who acted among other pro- 
minent citizens and tacksmen at Dornoch as juryman in a 
succession case. Sir John Matheson was Chancellor of 
the Bishopric from 1544 to 1554, and Robert Matheson, 
saddler and burgess of Dornoch (1566 to 1603), was a 
man of property, a fact to which we owe a list of the chief 
burgesses of Dornoch Burgh in 1603, who held an 
" inquest " over his estate. Colonel George Matheson of 
Shiness made his name on the Continent with Lord Reay ; 
and from him are descended the Mathesons of the Lewis 
— the late Sir James Matheson and Donald Matheson, the 
present proprietor of that island. 

The list of Burgesses of Dornoch referred to above 
must be given to show the character of the better class of 


the population, at least from the standpoint of names. 
They are these in 1603: — 

Richard Murray, 

Alexander Murray, 

William Murray M'Kane M'Kwatt, 

John Murray M'Kane M'Kwatt, 

Thomas Murray Angus-sone, 

Donald Mackphaill, 

Alexander M'Kraith, 

William Clunes, 

Alexander Clark, 

Thomas Veir, 

Thomas Ratter, 

George Dicksone, 

Thomas Fiddes, 

to whom add Robert Mathiescne, saddler, deceased. 
Earlier Dornoch names are : — Morays in plenty ; in 1512 
David Mudy (Moody), a family name introduced by Bishop 
William Mudy about 1450; in 1529, Ysaac Leslie and 
John Talyour; 1542, Walter Leslie, Ferquhard 
M'Gillespy, Alexander Rater and Alexander M'Culloch; 
1544, David Dyksoun ; in 1551 and 1552, in addition to 
these Leslies, Morays, M'Cullochs, and Dicksons, are 
Thomas Chesholme and John Gillepatrick Tailyeour; in 
1583 Angus Poison is a citizen, and the name Suther- 
land also appears, though rarely. The McRaes are also 
common as burgesses and once as bailie ; Robert 
McRaithe, bailie, 1570, with Angus McCraithe, holding 
Achloch in 1584, and Alexander McKraith, burgess in 
1603. Sir Robert McCraith was vicar of Kilmalie or 
Golspie in 1545. Other names connected with Dornoch 
City are Donald McGillemor (1512), Alexander Gar 
(G. GEARR, short) in 1545, John Awloche (Atholeman?) 
in 1524 and 1545, Thomas Mowate, Robert Duf or 


McDonald McDavid, 1562, John McDonald McMurquhe, 
1568, etc. The name Mowatt, so common in Caithness 
early and late, is in the oldest documents given in its 
original form of " de Monte Alto " — High Mount, whence 
the modern name is contracted and degraded. 


The clans outside Sutherland proper — those of Strath- 
naver, Durness, Eddrachilles and Assynt — did not in olden 
times comprise any great variety of names. This region is 
the cradle of one great clan — the Mackays of Strathnaver 
and Reay. The Mackenzies, with their dependants 
the Mclvers, and the Macleods, with the Morrisons 
and other Lewis septs, are outsiders, comparatively 
speaking. Like all the other northern clans — the Mac- 
leods, Mackenzies, Mackintoshes, Camerons, etc. — the 
Mackay Clan begins its history really at the end of 
the fourteenth century — at 1400. True, historians do 
tell us that Alexander Mackay, hailing from Galloway, 
the seat of the similarly named septs of Mackie and 
McGhie, or from Aberdeenshire and the lands of the 
Forbeses there, settled in Strathnaver in 1196, he having, 
of course, helped King WilHam to expel the Norsemen. 
Of course his son (or himself) married the Bishop of 
Cathanesia's daughter (as did the ancestor of the Morri- 
sons of Durness, etc.), and got from him Church lands in 
Strathnaver. Equally, of course, Magnus, the 4th from 
Alexander, fought for Bruce at Bannockburn, as did 18 
other Highland chiefs, inclusive of The Macpherson. The 
first assured chief is Angus Dubh Mackay, who in 1411 
barred Donald of Harlaw's path at Dingwall, and got well 
" thrashed " for the same. From him the descent can be 
followed easily to the present Lord Reay. The Mackay 
chiefs got their first charters only in 1499. 


Sir Robert Gordon calls them Clan Worgan — that is, 
Clan of Morgan — and though this title has been refused 
by the clan historian, under the impression that it was an 
invention of the enemy, it is the usual Gaelic name in 
literature for the Mackays of Sutherland, distinguishing 
them from the Mackays of Kintyre and " Mac Aoidh na 
Ranna," in Islay. In the famous " arming " piece in the 
Red Book of Clanranald they are called Clanna meram- 
oenmnacha, masgalacha, morbhrontach Morguinn agus 
Catuigh " — the merry-hearted, courteous, great-bestowing 
Clans Morgan and Cattach. This was written before Sir 
Robert's time, possibly. Niall McVurich says that Donald 
Duval Mackay O'Manus was " Morbhair Meghrath na 
chenn ar Mhorgannachuibh " — the Mormaor of Moyra 
(Lord Reay) was the chief of the Morgans. 

The name Morgan has puzzled and annoyed the his- 
torians ; it is Welsh, they thought. Now, the name is a 
good Pictish one, common in Aberdeenshire now, and 
especially in olden times, appearing in the roll of Earls of 
Mar and other dignities. It is also in the Book of Deer 
(circ 1100). It is the old Celtic name Moricantos, " Sea- 
bright." Its preservation in Aberdeenshire and Strath- 
naver is interesting ; and the fact shows that there is more 
in the Forbes myth than some wise historians think. The 
name Mackay is, in Gaelic, Mac Aoidh, son of Aodh, and 
this in old Gaelic was Aed, the Celtic Aidus, which was the 
word, declension and all, for " fire." That it was once a 
longer name — such as Aed-gal, Aed-gin, Aed-lug, etc. — is 
possible ; but in historic times it has been Aed, and means 
" fire," neither more nor less. Caesar's Aedui, whose 
name is directly from Celtic AEDUS or AIDUS, were the 
first Mackays! 



I have no intention of dealing with the purely 
physical antiquities of Sutherland, such as the cairns and 
stone circles of the Stone Age, or the brochs (of which 
there are 60) and earth-houses of the Bronze and Iron 
Ages. I will deal only with the literary monument left 
by early Christianity at Golspie in the shape of an Ogam- 
inscribed monolith usually known as the " Golspie Stone," 
now in the Dunrobin Museum. The inscription on this 
stone has received much attention in late years from Mr 
Nicholson, of the Bodleian Library, and from Professor 
Rhys. Both have written largely on the so-called 
" Pictish Inscriptions " in the Ogam character. The Ogam 
letters are an Irish invention — a sort of proto-telegraphic 
system where the letters of the alphabet are denoted by 
so many strokes — from one to five — above, through, or 
under a stem line respectively. The letters are easy to 
inscribe, but often difficult properly to read. 

Professor Rhys has spent many years in deciphering 
these Ogam monuments, but Mr Nicholson " came, saw 
and conquered " all at once. As he says himself, it was 
on a visit to Golspie in 1893 he came " by chance " to 
study the Pictish inscriptions of Scotland, and two years 
later he gave his " chance " lucubrations to an astonished 
Celtic world in the pages of the " Academy." Since then 
he has put the articles together in book form under the 
title of the " Vernacular Inscriptions of the Ancient 
Kingdom of Alban." Celtic scholars have not thought it 
worth while to confute Mr Nicholson's views ; his philology 
belongs to the good old days of Charles Mackay and 
Lachlan Maclean (" Lachunn nam Mogan.") The Picts, 
according to Mr Nicholson, were Gaelic-speaking Celts. 
Now, no Celtic scholar holds such a view ; even Professor 
Rhys maintains that they were not Gaelic-speaking. In 


his extraordinary paper in the Society of Antiquaries' 
Transactions (1892), and lately in the report of the Welsh 
Land Commission, the Professor strenuously holds that 
the inscriptions are written in a lost language, which 
was neither Gaelic nor Welsh, nor allied to them at all. 
Dr Whitley Stokes and Professor Windish consider the 
Picts to have been of Brittonic race and language, and 
the place-names of Pictland alone ought to be enough to 
bring any unprejudiced mind to this view. It is, however, 
probable that the inscriptions are in Gaelic, for they were 
no doubt the work of the Gaelic-speaking missionaries 
from Ireland who Christianised Pictland. The Ogam 
writing is, as already said, a purely Irish invention. The 
Ogams are, therefore, not " Pictish ;" they are the " Ogams 
of Pictland," and " Pictish inscriptions " is a misleading 
term. Mr Nicholson's greatest sin is his disregard of 
what Professor Rhys calls " perspective in language." He 
explains Tth century Gaelic as if it were 19th century 
Gaehc. He forgets or overlooks the fact that " b " and 
"m," for instance, w^ere not aspirated for hundreds of years 
after the Ogams were inscribed on the Golspie stone. 
The readings of these Ogams are unsatisfactory in the 
highest degree. Professor Rhys pins his faith to an atro- 
city like this at Lunasting — xttocuhetts : ahehhtmnnn : 
hccwevv: nehhtonn. He challenges the believers in the 
Brittonic origin of Pictish to explain it, and says — " Let 
them explain it as Welsh and I shall have to confess that 
I have never understood a single word of my mother 
tongue." ! ! ! As it stands printed above, it belongs to 
a language that " was never heard on land or sea." It 
looks like the language of Luna, the moon — " Lunacy," it 
may be named. 

Mr Nicholson's reading of the Golspie stone is equally 
as satisfactory as Professor Rhys's champion inscription 
given above. It runs thus — Allhhallorr edd m'qq Nu 
uvvarrecch. This he puts in Gaelic thus — Alhallr, ait Mic 


Nu Uabhraich, " Alhallr, the place of McNu the Bold." 
He takes Alhallr to be Norse, meaning " All-sloping." 
The stone — indeed, all the inscribed stones — he regards 
as boundary stones, not gravestones. McNu is compared 
to the name of Columba's maternal grandfather Mac Naue, 
a perfectly legitimate comparison, and to the Pictish 
McNu Sir Herbert Maxwell correlates the Galwegian and 
Dumfries names of McNoe or McNoah. Professor Rhys 
reads the stone thus — Allhhallorr edd Maqq Nuuwa 
rreirng. It is at present useless to speculate on the mean- 
ing of such a conglomeration of letters ; the stone is evi- 
dently mis-read, and so are most of them. 

A name about which much nonsense has been written 
is Dunrobin ; and the latest nonsense is in Mr Nicholson's 
book on " Golspie." If the name had been Dun-Robert, 
he says, it would be easily seen that it was named after 
some Robert or other; but Robin he regards as a name 
unlikely among a Gaelic population. This is a fallacy. 
Diminutives like Robin in -in and -on were common in 
early times with English names borrowed by the Gaels, 
and in the Highlands we meet with many — Gibbon 
(whence McGibbon), Robin (whence the present-day 
McRobin), Paton (whence AlcFadyen;, Wilkin (whence 
McCuilcein), Rankin (whence McRankin, now obsolete), 
Cubbin (from Cuthbert, whence McCubin, a Kirkcudbright 
name), Michin (from Michael), whence McMichin or 
McMeeking). In Ireland such forms were also common 
— Tomin, Wattin, Philbin, Rickm, Robin. Dunrobin, 
then, means the DUN or " fort " of Robin. The form 
Drum-robin also occurs: " Robin's-ridge." Dunrobin 
appears first in 1401 in Earl Robert's charters; it is no 
doubt named after him. 





THE Island of Skye is first mentioned by Ptolemy, 
the classical geographer of the second century. 
He calls it Skitis or Sketis, and misplaces it, putting 
it where the Orkneys should be. The island is 
next mentioned by Adamnan (700 A.D.), who calls 
it Scia Insula, the Island Scia, no doubt from Scith. St 
Columba, he tells us, was there once baptising a pagan 
chief called Artbrananus at a well called after him, 
Dobur Artbranani, at another time losing himself in a 
dense wood, for the island, we know, was once well 
wooded, as, indeed, were all the Hebrides till the advent 
of the Norse. The Irish annals tell us that in 668 the 
sons of Gartnait, with the people of Scith (or Sceth, gen. 
case), migrated to Ireland, but returned two years later. 
In TOO A.D, a battle of some consequence took place in 
Scii (dat. case), followed by the destruction of Dunolly 
Castle by King Selbach. Skye is mentioned several 
times in the Norse sagas, especially as being devastated 
in 1098 by Magnus of Norway : the Norse called it Skith, 
which in their language meant " tablet, log ;" and, evi- 
dently from a remembrance of the Norse possession of 
the island and its Norse name, our poets have called the 


isle by the name of Clar Sgithe. The Dean of Lismore 
(1512) speaks of 

" McWllam oo Clar Skeith," 

the Dunvegan chief of the time. Rory M'Vurich's elegy 
on M'Leod (pubhshed 1776) has it— 

" Dh' fhalbh mo lathaichean eibhinn 
C' 'n threig sibh Clar Sgithe." 

The early charter forms of the name are Skey (1292), 
Sky (1336, the 14th century form), Skye (1498) ; in the 
Chronicle of Man the name is Ski. The present Gaelic 
name is adjectival: An t-Eilean Sgiathanach, and Sgith- 
eanach, explained as the " Winged Isle." The older forms 
of the name, Sci or Scith, point to a shorter form of the 
same sort, as in Gaelic SGIATH, Norse SKITH, namely SKI, 
divide. The name may really be Pictish and mean the 
" Indented Isle," still having the same general force as 
" Wmged." 

It is usual to connect with Skye the early history 
of Cuchulinn, who received his martial education in the 
isle and DUN of Scathach, the Amazon Championess. The 
oldest tales make her live in an isle eastward of Alba or 
Scotland ; the Scottish tradition has it that Dun-Scathaich 
in Skye is meant (Dunskahay in 1505 — Dun-Scathaigh', 
the Dean of Lismore's Zown Skayth, and Skay for dative 
of Scathach), and this, after all, is likely. 

Skye was one of the immediate causes of Hacon's 
invasion in 1263 : with the rest of the Isles it owned 
Norway as its suzerain, but it formed part of the Kingdom 
of Man and the Isles till 1266, when the cession of the Isles 
took place. The Earl of Ross attacked it fiercely in 1262, 
with the consequence that complaint Vv-as made to Norway. 
Hacon in 1263 passed between Skye and the mainland 


with his fleet, and there is httle doubt that Kyleakin — 
" Hacon's Sound," as it undoubtedly means — got its name 
from the passing through of Hacon and his fleet. Simi- 
larly King James IV.'s punitive visit to Skye and the Isles 
in 1540 is no doubt remembered in Portree, the King's 
Port. The Earl of Ross received Skye as his portion of 
the spoils of Largs and the events of 1262, and in 1292 
the Sheriffdom of Skey was constituted from the Earl of 
Ross's lands on the West Coast and northern Hebrides 
(Skye and Lewis especially), for which Hugh of Ross has 
in 1309 a charter from Robert Bruce. In 1335 King 
Edward Balliol forfeited the next Earl and gave the Isles 
of Skye and Lewis to John of Isles, the head of the Clan 
Donald ; but David Bruce, on his return, restored Skye to 
the Earldom of Ross. It passed, with the Earldom of 
Ross, into the hands of the Wolf of Badenoch, the LesHes, 
and finally Donald of the Isles took it with the rest of the 
Earldom of Ross. The forfeiture of the Lord of the Isles 
towards the end of the 15th century brought the local 
clans into direct relations with the Crown, and hencefor- 
ward we have Macleods, Macdonalds, and Mackinnons as 
leading chiefs and landlords in Skye. 

The completeness of the Norse conquest and posses- 
sion of Skye is proved by the place-names. The seven 
natural provinces of the island have all Norse names — 
Trotternish, Waternish, Duirinish, Bracadale, Minginish, 
Strathordil, and Sleat ; and the important townships and 
" setters " also were occupied and named by the Norse- 
men. The Valuation Roll is the best proof of this. Some 
250 different names occur in Skye in the present Roll, but 
nearly a score of these are purely English. The rest are 
Gaelic and Norse, and the proportion is exactly 60 per 
cent. Norse names and 40 per cent. Gaelic. Naturally the 
percentage is heaviest on the west and north, while Sleat 
and Portree have the fewest Norse names. Black's tourist 
map of 4 miles to the inch has 58 per cent. Norse names 


for Kilmuir and 33 for Sleat; the latest map — 2 miles to 
the inch — gives 55 per cent. Norse names for Kilmuir, and 
Sleat 30 per cent. It practically contains all the names in 
the ordnance one inch to the mile map. As Duirinish and 
Snizort show a higher percentage on the Valuation Roll 
respectively than Kilmuir (viz., 68, 70, 61 per cent.), we 
may roughly state that about 50 per cent, of the names on 
the 1 inch map are Norse as against the Gaehc names. 
Of course, as the maps increase in minutiae and detail, the 
Gaelic names correspondingly increase : the smaller fea- 
tures — fields, hillocks, and burns — are apt to be Gaelic, 
having possibly been re-named since the Norse occupa- 

Of the present parish names three are Gaelic — 
Strath, Portree, and Kilmuir (St Mary's kil or church), 
but the first is properly half Norse, being for Strath- 
Swordale or Strathordil, where Swordale or Suordale is a 
common Norse name, appearing otherwise in Skye (at 
Dunvegan), in Lewis, in Ross-shire, and in Sutherland, 
explained by Captain Thomas as " Sward-dale," which it 
likely is, though Saur-dalr, " Swampy or Mud dale," may 
suit the phonetics in this case. The other four parishes 
have Norse names : Duirinish, Hke Durness, in Sutherland, 
means " Deer's ness," NESS being the Norse NES, a cape, 
nose; Snizort, Snesfurd 1501, Sneisport 1526, Snisport 
1561, seems to be for Snaes-fjord, "Snow-firth," the sug- 
gested " Sneis firth," from the name Sneis (skewer) being 
unsatisfactory. Bracadale (Bracadoll 1498, Braikodell 
1541) stands for Brekka-dalr, " Slope-dale ;" and Sleat is 
the Norse sletta, a level or plain, it being, in fact, the most 
even or level portion of Skye. Here we may add the dis- 
tricts of Trotternish, Waternish, and Mmgmish. The first 
is spelt by Dean Munro (1549) as Trouternesse or Tron- 
ternesse, and a writer at the end of the 17th century also 
spells the name with N, not U, before the middle T. 
M'Vurich gives the Gaelic for it in the Red Book of Clan- 


ranald as Trontarnis ; the modern Gaelic is practically the 
same — Tr6(n)dar-nis, with nasalised 6. It stands for 
Throndar-nes, " Thrond's ness," Thrond being a favourite 
personal name, giving rise also to many place-names, in- 
clusive of the Icelandic Throndarnes quoted above, and 
the modern Drontheim or Trondhjem, in Norway. 
Waternish finds its counterpart in two Icelandic nesses 
called Vatnsnes, " Water-ness ;" while Minginish 
(Myngnes, 1498, Mygnes 1511, Myngynnes 1549, I\Ien- 
zenise 1549) may stand for Megin-nes' " JMain ness." 

We may next deal with the islands belonging to Skye. 
Pabay is for Pap-ey, " Pope or Priest Isle," a common isle 
name, showing the Norse found the Gaelic hermits or 
Culdees there. Scalpa, which also appears in Harris, is 
found also in Orkney, but there stands for Skalp-eith, 
" Ship isthmus," whereas in the Hebrides it means " Ship 
isle." Raasay (Rasay 1501, Rairsay 1526, Raarsay 1549) 
seems to stand for Rar-ass-ey, " Isle of Roe-ridge." Its 
neighbour, Rona, is easily derived, for we have the Norse 
name for it — Rauney, that is Hraun-ey, " Isle of the rough, 
rocky surface " (HRAUN, lava-field especially), a deriva- 
tion supported by Martin's description of it — " This little 
isle is the most unequal rocky piece of ground to be seen 
anywhere." Staffm Isle is so named from its basaltic 
rocks, Norse STAFR, staff, etc. ; it gives its name to the 
adjoining bay and mainland. Altavaig, on the contrary, 
must take its name from Alpta-vik, " swan-bay." Trodday, 
to the north of Trotternish, may be Thrond's Isle, though 
" Pasture (Trodh) Isle " has also been suggested. Holm 
Isle, off Duntulm, gives its name to the latter place, " Fort 
of Holm ;" the Norse is HOLMR, a holm or islet in a bay. 
The Ascrib Isles, called by Monro Askerin, seem to con- 
tain a reference to ASKR, ash, spear, ship. Isay is " Ice- 
isle ;" Mingay, " Lesser-isle ;" Wiay, a common isle name 
in the Hebrides, seems for Ve-ey, " Temple-isle." Oronsa, 


off Bracadale and off Sleat (the latter Eilean Dhiarmaid ; 
ITth century Island Diermand), is for Orfiris-ey, " Ebb- 
tide-isle," joined to the mainland at low water. There are 
three or more other such. Soay, also a common island 
name, " Sheep-isle," as Lampay, at Dunvegan, means 
" Lamb-ey," and Eilean Heast, or " Horse-isle," no doubt 
gives its name to the mainland township of Heast, though 
in Iceland a place gets its name Hestr on account of a 
horse-shaped crag. 

The sea-lochs or fjords have mostly Norse names end- 
ing in ORD or ORT, derived from fjord. Broadford, like 
the Arran Brodick, has replaced a Norse Breidha-fjord. 
Lochs Ainort and Eynort may be for Einar's fjord ; Loch 
Snizort has been already explained ; Loch Harport is pos- 
sibly Hafra-fjord, " He-goat " fjord ; and Loch Eishort 
may be for Eiths-fjord, or Isthmus fjord. The termina- 
tion AIG stands for VIK, bay, and there are many such. In 
Sleat we have Ostaig, " Eastwick ;" Saasaig, Cask 
bay (sas-vik) ;" Morsaig, "Ant-bay (Maurs-vik") ; Aula- 
vaig, " Olave's bay ;" Tarskavaig, "Cod-bay (thorskr, 
Gaelic trosg ") ; in Strath, Malag "Measure or Speech bay 
(Mala-vik ") ; Boreraig, " Burg bay " (Borgar-vik ") ; in 
Bracadale, Scavaig river, ITth century Scah-vag, " Shaw- 
bay ;" Fiskavaig, " Fish-bay ;" Totaig, " Toft or Clearing 
Bay ;" in Duirinish, Varkasaig, " Castle-bay " (ViRKi) ; 
Branderscaig, Ramasaig, " Raven's bay ;" Boreraig and 
Totaig as before, Camalaig ; in Snizort, Pen-soraig, from 
Saur-vik, " Mud bay ;" Liuravaig, like Lerwick, " Mud- 
bay ;" Bearraraig, Bjorn's bay ; in Kilmuir, Bornaskitaig 
(Martin's Bornswittag), from Skipta-vik, " Division-bay ;" 
Volovig, " Field bay ;" Loch Langaig, " Long-bay ;" Bro- 
gaig, " Breeches' bay ;" Torvaig, " Thori's bay ;" Cracaig, 
" Crook-bay ;" Tianavaig (Martin's Camstinvag), Tmdar- 
vik or " Peak-bay ;" Oskaig (long o), m 1630 Oistage, 
" Osk's bay or the Desire bay." A river mouth is OSS 
or AR-OSS, and hence come Ose, Glen-ose, Osdal, Oisgill 


(Oyce-burn), Aros Bay, Inveraros or Inverarish in Raasay. 
The capes or nesses arc numerous : Ard Thurinish, 
" Thwart or cross (Thver) point ;" Ardnish, a hybrid, as 
also Kraiknish, UlHnish, " UlH's point ;" Crossnish, " Cross 
point ;" Uignish, " Wick ness ;" Unish, " Yew ( ?) ness ; 
Greshornish, " Grice or Pig's ness ;" Skirinish (Martin's 
Skerines), " Skerry ness ;" Meanish, Mjo-nes or " Narrow 
ness ;" Hunish, " Bear's (hunn) ness ;" Arnish " Erne or 
Eagle ness;" Manish (Maenes, 1630), "Sea-mew ness." 
The two south capes of Suisnish — Raasay and Strath — 
come from SNOS, projecting rock, head, in fact for " Sow's 
head." Eyrr, a pebbly beach, appears in Eyre of Snizort, 
Ken-sal-eyre, and Ayre Point, south of Raasay, 

Hill and rock names are mostly Norse, showing VALL 
for FJALL, fell; CLEIT or KLETTR, a rock; STACK or 
STAKKV, a stack of a hill ; SGEIR, skerry ; SCORE and 
SGUVR, edge or chff ; HAMARR, rock (as in Hamara, " Rock 
water ") ; SGATH or SKAGI, a jutting hill or promontory 
(Beinn a' Sgath) ; and Sco, a shaw or woody rising 
(Birkisco, " Birk-shaw.") The Hoe and Cop-na-Hoe 
mean HAUGR, tumulus or burial hillock. Ben Cleit, Loch 
Cleit and such names are common. Stockval is " Stock- 
fell ;" Arnaval, " Erne-fell ;" Roineval, " Hraun-fell," as in 
Rona; Scoval, "Shaw-fell;" Horneval, "Horn-hill;" 
Helavall, " Hella or Flagstone fell;" Maehall, "Narrow- 
hill ;" Reiveal, " Smooth-hill." Ben Storr is from STOR, 
big. To these must be added one or two names in BRECK 
or BREKK, a slope ; Cross-breck and Scorry-breck. 

All names with -DAL as an ending are Norse, and 
they are very numerous (Armadale, Meadale, etc.) There 
are many GILS or burns (Vidi-gill, Vikisgill, etc.); but 
Idrigill, which occurs twice, is a promontory, and must 
stand for Ytri-kollr or " Further (Outer) hill." There are 
three names from FORS, a waterfall — Forse and Forsan. 


The township names show many -SHADERS ter- 
minally ; this is the Norse SETR, a station or sheiling 
(Marishader, " mare-seat," Shuhshader, " Pillar-seat," etc.) 
The suffix BOST means township — Husabost, " House- 
stead ;" Breabost, " Broad-town," etc. Terminal STA is 
for STADR, a stead, holding — Shuhsta, as in Shulisshader, 
Lusta, Conista (Lady's-town), etc. Suffix GARRY is for 
GARTH — Bigeary, "Bigging-garth," etc. BuRG or Borve 
is very common — Skudiburg, Raisaburg, etc. ; so is TOT or 
TOBHTA, a toft or clearing. 





THE earliest mention of Lochaber occurs in the pages 
of Adamnan, who died in 704. St Columba, he 
tells us, gave a miraculous stake, which attracted 
and killed wild animals coming near it, to a poor 
man from " the district which borders on the shores 
of the Apcric Lake ;" and the poor man further catches 
by means of the stake a monster salmon from the river 
there — a river " which may be called in Latin Nigra 
Dea," or Dark Goddess. Dr Skene has ingeniously con- 
nected this river with another name given by Adamnan 
for a lake in the Drumalban Range, viz., Loch-dae, which 
means in old Gaelic " Dark Goddess," from obsolete 
LOCH, dark. Hence the Nigra Dea is the Lochy, either 
in Perth or Lochaber ; and it is an interesting reminder of 
the fact that the ancient Celts v/ere great river wor- 
shippers, which they often called " Mothers " (Matrona 
or Marne), or " Goddesses " (Deva or Dee, Divona or Don). 
Adamnan's Aporic comes from the old Gaelic APOR, a 
marsh), and the meaning of Lochaber is the Lake of the 


Marsh. Fortunately, tradition supports this view, for, 
according to it, the original Lochaber was a lakelet in the 
Moine Mhor — the Large Moss — near the mouth of the 
river Lochy. Tradition and philology thus go hand in 
hand in the etymology of the name Lochaber : no scep- 
ticism need apply. The old Gaelic APOR and Irish ABAR 
(marsh) are not the same as the Pictish prefix ABER or 
OBAIR (confluence, river-mouth), so common in place- 
names in Pictland, such as Aberdeen, Aberchalder, and 
the like. Indeed, only one characteristically Pictish name 
can be fomnd in Lochaber, and that is the old Pitmaglassy 
(1500), or Pittenglass (1669), which is mentioned along 
with Achadrom and Culross, on the borders of Lochaber 
and Glengarry, but such names as Spean or Pean may be 
inferred to be Pictish names belonging to the main fea- 
tures of the country. It is clear that in Columba's time 
the inhabitants of Lochaber spoke Gaelic, not Pictish: 
they, in fact, belonged to the old Scottic invaders of Pict- 
land that settled in Oirir-Ghaidheal (" coastland of the 
Gael," Argyll from Kintyre to Lochbroom) long ere the 
Sons of Ere founded the Kingdom of Dalriada. Hence 
the people of Lochaber speak the Northern Dialect of 
Gaelic, and good Gaelic it once was, and still is. 

Romance makes Banquo of the Macbeth drama Thane 
of Lochaber, and respectable tradition connects the Cum- 
mings with Lochaber in the thirteenth century ; and, of 
course, as elsewhere, there is a Cummings-tower at Inver- 
lochy Castle. When we come to assured history, we find 
that the " lands of Louchabre " formed part of the Earl- 
dom of Moray, granted to Randolph between 1307 and 
1314, though seemingly it soon left his hands, for in the 
" Index to Lost Charters " it is entered that Angus of the 
Isles got the lands of " Lochabre " from Bruce in 1309. 
Anyway, Lochaber was not long attached to the Province 
of Moray. The " good John of Isla " got the ward of it 
in 1335 in the minority of the Earl of Atholl, who pro- 


bably got it through his grandmother, the heiress of the 
Cummings. The Island Lord got it altogether in 1343. 
Alaster Carrach, third son of the good John of He, is in 
possession of the Lordship of Louchabre at the end of 
the century, though the superiority still remained in the 
hands of the eldest son, the Lord of the Isles. Alaster 
Carrach was the ancestor of the Keppoch Macdonalds, and 
it was from him, as eldest direct descendants, that their 
claims of Brae Lochaber arose. The last two Lords of the 
Isles seem to have ignored Alaster Carrach's descendants ; 
at anyrate, in 1444, Mackintosh got from the Lord of the 
Isles the lands of Glenluy and Locharkaig, west of Loch 
Lochy, and the Brae Lochaber lands, at first as far down 
as Loch Lochy, the chief demesne being Keppoch. Mac- 
kintosh's claims to these lands and his attempted posses- 
sion of them as against the native Camerons of Glenluy 
and Locharkaig and the Macdonalds of Keppoch, kept 
the country in a state of turmoil for two hundred years. 
Finally, in 1666, Lochiel unwillingly bought Glenluy and 
Locharkaig, Argyle being guarantor ; but the Keppochs 
kept forcible possession longer, which practically ended in 
the last clan fight in Scotland, on Mulroy, above Keppoch, 
in 1688. Though then defeated. Mackintosh made good 
his right, and still holds the property. 

The Camerons also incurred the displeasure of the 
Lords of the Isles for deserting them in 1429 in their 
quarrel with the King. Their brave and energetic chief, 
Donald Dubh, fought against Donald Balloch at Inver- 
lochy in 1431, when the latter defeated the Royal forces. 
Donald Dubh and his clan suffered severely over this, and 
became somewhat Ishmaelite for a century or two there- 
after. Maclean, first, of Coll, then, of Lochbuy, received 
the barony of Lochiel, there being a charter to the latter 
in 1461 of these lands. The Lord of the Isles' cousin, 
Celestine of Lochalsh, who appears to have received the 
lordship of Lochaber from the Earl, favoured Cameron as 


against the Macleans, and granted him Lochiel in 1472 — 
a grant renewed in 1492 by Celestine's son. The final for- 
feiture of the Lord of the Isles in 1493 brought the lord- 
ship of Lochaber into the King's hands. In 1494 Mac- 
lean again got Lochiel, but next year Cameron received 
it from the Crown. The Macleans kept up their claims 
or sold them, and the frequent forfeitures of Cameron gave 
his enemies plenty of opportunity to get charter rights to 
his property. After passing through the hands of Mac- 
leans, Campbells of Cawdor and Argyll, and Huntly, it is 
gratifying to record that the Chief of Clan Cameron still 
holds the barony of Lochiel — the land north of the loch 
and west of Loch Lochy. 

The portion of Lochaber between Glen Nevis and Loch 
Leven, east of Loch Linnhe, had the general title of Maw- 
more, the "Great Mam," MAM meaning a "large round hill." 
The name is still preserved in Mamore Forest. Maw- 
more was still in the King's lands after the forfeiture of the 
Lord of the Isles in 1502, but then it was granted to 
Stewart of Appin, and in 1504 to Huntly, who as Lord of 
the Lordship of Lochaber sold it in 1522 to Argyll. 
Mamore came thereafter into the hands of the Gordons, 
but on their decline the western half of it came into the 
hands of Lochiel, and the eastern half, inclusive of Glen- 
Nevis, belongs to Mrs Cameron-Campbell of Monzie repre- 
sentative of the Camerons of Fassifern. Glen Nevis of 
old belonged to the Camerons. The district about Fort- 
William and Inverlochy, now the Abinger estate, was 
acquired by Maclean of Duart in 1496, who sold to Stewart 
of Appin and to Huntly parts of the estate, and in 1531 
Lochiel got Appin's lands, including Inverlochy. But the 
Duart family again recovered the land (1540), and in the 
ITth century it finally came into Huntly's hands, who held 
it in the 17th and 18th centuries. The Gordons parted 
with the estate some fifty years ago, and it is now Lord 
Abinger's. The leading proprietors of Lochaber are 


therefore Cameron of Lochiel, Mackintosh of Mackintosh, 
and Mrs Campbell of Monzie — three good Gaelic family 
names as against one "Scarlett" Sassenach, who holds a 
big fourth of the country ! 

Lochaber now comprises the two parishes of Kilmallie 
and Kilmonivaig; but formerly Mamore formed part of 
the old parish of Elan-munne, the other part belonging to 
Argyll. The name Elan-munne, or St Munn's Isle, is 
still known, corrupted in English to Mungo. It is an 
islet in Loch Leven, where St Munn's Church once was, 
and the burial-ground still is. Munn, or Munna, is a con- 
traction for Mo-Fhindu, "My Findan," and is a pet name 
for St Finnan. Kilmonivaig, in 1449 Kilmanawik, in 1500 
Kilmonyvaig, stands for Kill or Church of St Mo-Naomhoc, 
or St Naomhan. Kilmallie appears in 1292 as Kilmalyn, 
in 1532 Kilmale, and 1552 Culmaly. This was also, spell- 
ings and all, the old name of Golspie. A Kilmayaille is 
mentioned in Tarbert of Kintyre. The name is a difficult 
one at best. There is no St Malli — the A is long — though 
there is a literary Charles O'Malley from an ancestral 
O'Mailli. Possibly our saint here is Amhalghaidh or 
Aulay, confused with Norse Olave (whence M'Aulay), 
with the usual MO prefixed. Kilmallie would then stand 
for Cille-mo-Amhalghaidh. 

In examining the place-names we may begin with 
the great natural features, the lakes, rivers and mountains. 
Loch Linnhe is known in the locality as An Linne, " the 
Pool " or " Sea-loch ;" the Lochiel part is called An Linne 
Dhubh, " the Black Pool," the part outside Corran An 
Linne Sheilich, " the Willow Pool." Lochiel, in 1461 
Locheale, Lochheil in 1492, and Lochiel in 1520, still 
awaits explanation. Loch Lochy has already been 
explained as the lake of the river " Dark Goddess." It 
must be remembered that lakes and glens take their names 
from the rivers that flow from or through them, a fact noted 


by Adamnan in speaking of Loch Ness, which he calls 
the " Lake of the River Ness." Loch Arkaig is named 
from the River Airceag ; its etymology is unknown. The 
root ARK means " defence " m Celtic ; but possibly the 
root here is ERC, darkish, the " dun " river being perhaps 
the force of the word. Equally enigmatical is Loch Treig 
and its river Treig, where the E is long. In modern 
Gaelic Treig means " forsake." Loch Eilde means the 
River and Loch of the Hind ; Loch Gulbin, Torgulbin, 
etc., come from GULBAN, a beak. Loch Leven and the 
River Leven, Gaelic Li'un, point to a Celtic Li-vo-na, 
root LI, flow, Gaelic LIGHE, a flood. There is a Leven in 
Lennox and another in Glen-Lyon similarly pronounced. 
The name is Pictish originally, as, indeed, may be the 
other difficult names above passed in review: perhaps 
older still. 

In regard to the bens and glens, the latter follow the 
river names, and the former are nearly all easily derived 
by anyone that knows Gaelic. Ben Nevis is the excep- 
tion. There we have Glen-Nevis and the River Nevis. 
A sea loch in the south of Knoydart is called Loch Nevis, 
and hence we are sure that it is the river that has given 
its name to Glen and Ben Nevis. The local pronuncia- 
tion is Nibheis (Eng. Nivesli), which would give a primi- 
tive form, Nibestis, but probably Nebestis is the right 
form, from NEB, burst, flow; the name may have been a 
goddess name allied to " Nymph " by root and idea. 
Glen-Spean, of course, takes its name from the River 
Spean, and the SP initial is hardly ever native Gaelic — ■ 
certainly not old Gaelic. We may conclude that Spean is 
Pictish, originally Spesona, the root SPES, SPE or SQES, 
SQE, as in Spey, to cast forth, squirt, vomit, Gaehc SGEITH ; 
and the old ItaHan river name, Vomanus (root VOM), may 
be instanced as analagous. The idea is that of a fast- 
flowing, " spatey " river, the word " spate " being also dis- 
tinctly aUied. Glenfmtaig comes from the River Fintag, 


the " white river," a common river name. Mrs Mackellar 
explained Glende^sary (Glendessorach, 1505) as the 
" Glen of the South Shealmg," DEAS-AIRIGH : anyway the 
root is DEAS, right or south. Glen-pean (so in Blaeu, 
Glen-pona, 1505), Glen-kingie (Glen-Kinglen, 1505), and 
Glen-gloy must be left alone at present ; Glenroy is the 
" Glen of the Red River ;" Glen-sulag, from Suileag river, 
the river full of " eyes " or pools ; and Glen-luy (Glenloy, 
1505, Glenlie, 151G), " Glen of Calves." 

Fort-William gets its name from the fort built for 
Wilham of Orange near there in the end of the seven- 
teenth century, but the village was originally named after 
his consort, Mary, hence Maryburgh. A hundred years 
earlier (159T) an Act of Parliament directed the creation 
of three burghs in the West Highlands to maintain the 
cause of " civility and policy ;" these were to be in Kin- 
tyre, Lochaber, and Lewis. The final result came to be 
the erection of the towns of Campbeltown, Fort-William, 
and Stornoway, but only the former procured the benefits 
of the Act in becoming a Royal Burgh. The Gordons, 
who owned the Inverlochy estates till their demise with 
the last duke (1836), changed Maryburgh into Gordons- 
burgh when the fit of " Orangeism " was past, and the 
village was last century so known. Sir Duncan Cameron 
of Fassifern, in succeeding the Gordons, changed the 
name to Duncansburgh ! This has also given way to the 
" garrison " name of Fort-William, it being An Gearasdan 
in Gaelic. Auchintore-beg (Auchintor-beg, 1496) was its 
pre-village name, the field of the TODHAR or manure. 

Only a few of the leading farms and other such names 
can here be taken. Callart, so in 1522, is explained as 
Cala-ard, the " high bay " literally, but " upper ferry " 
really, as opposed to the Ballachulish one. Ballachulish 
(Ballecheles, 1522), " Straits town," Kinlochmore, Cor- 
uanan (" Lamb's Corry,") Blar-nan-cleireach, and such 


in " Mamore " are easy. Onich, Offanych in 1522, is, as 
its older form proves, from OMHANACH, " foam-frothed " 
place ; Cuilchenna is shown by the old form, Culkenan 
(1522), to mean " Back of the Head-ie ;" Corran means a 
small promontory between two bays, not a " sickle," as 
usually explained ; Drumarbin, Blaeu's Druimerbin, Drum- 
arbane of 1522, is no doubt " Ridge of Roes." Lundavra 
is in 1502 Dundabray, a fort or forts in a small isle in 
Mamore, explained by Mrs Mackellar as Dun-da-rath„ 
" Fort of two RATHS or enclosures," there being two islets. 
Mr Livingstone explains the name as wave (lunn) of the 
double crest (bra) ; but Lundavra seems a phonetic assim- 
ilation arising from Loch Dun-da-bhra. Dun-dearduil is 
etymologised in the Ordnance maps into " Fort of the 
Foreign Bard!" The Dearduil here is a common fort 
name, usually explained from the heroine Deirdre, Mac- 
pherson's Darthula. 

In Kilmonivaig the most interesting farm name is 
Keppoch; it is from the old Gaelic CEAPACH, a tillage 
plot, a holding, in root the same as CEAP, block. Fersit, 
the "two Fersenas" of 1500, and Blaeu's Farset, comes 
from FEARSAD, an estuary, sand-bank, which suits the 
place. Names like Blarour, Tirandrish, Achluachrach, 
Achaderry, Tulloch, Innerroy, and Innis or Inch have 
only to be pronounced in GaeHc and their secret is out. 
Lianachan is from LEAN, a mead ; Sliabh Lorgach, east of 
Loch Treig, gets its name from its moraine " tracks " or 
lines; Bothuntin, Gaelic Both-hundainn, Bothinton in 
1444, is one of the many BOTHS or " steads " of Brae- 
Lochaber, and it is a puzzhng name, for Hundaidh or 
Hundainn did duty in Gaelic for Huntly ; only here the 
appearance of the name in 1444, 1466, and 1476 precludes 
connection. It is evidently the same name as we have in 
Contin, a " confluence " place. 


In Kilmallie we may note Corpach, so named, as the 
Old Statistical Account says, from the fact that here they 
rested with the bodies before embarkation on their way to 
lona. Annat comes from the obsolete ANNAID a mother- 
church, doubtless where the anchorite had his cell and little 
chapel. Fassfern is the " abode " or FASADH of the alder ; 
Errocht means in old Gaelic a " meeting ;" Moy, a " plain ;" 
and Glastor is etymologised correctly on the ordnance maps 
•Glas-doire. Other Lochaber names of interest — and there 
are many more such — are worthy of record and study. 





LOCHALSH is undoubtedly the Volsas or Volas Bay 
of Ptolemy, the geographer of the early part of the 
second century. It is therefore one of the oldest 
names in Scotland. Its next appearance in written 
form is more than thirteen centuries later: this is 
in 1464, in a Crown charter to Celestine of the Isles, where 
the name is spelt Lochalsche. A common form in early 
writings was Lochalch or Lochelch, and then Lochalse 
(1576), now Lochalsh (Gaelic Loch-aills' or even Loch-ais', 
where the L, as usual, drops before the S. Probably the 
form Volsas is the ancestor of our modern form ; Volas 
might be apt to land in Loch-aill, whereas Volsas would 
give, according to well-known phonetic laws, latterly 
Fallas or such. That root may be VOL, roll, as a wave ; 
Eng., WELL. 

Lochalsh formed part of Argyle, which anciently 
stretched as far as Lochbroom ; it belonged to North 
'Argyle, the present county being South Argyle. It also 
belonged to the Earldom of Ross. It was the " patr-ia," 


or habitat, of pre-Mathesons, and the TOISEACH, or thane, 
of Lochalsh — the Chief of the Mathesons — was one of the 
Earl of Ross's leading vassals. Similarly Kintail was no 
doubt held by a TOISEACH, who latterly was Mackenzie of 
Kintail. The Edinburgh MS. (1450) indicates four clans 
in early times (1200-1400) as inhabiting from Lochalsh to 
Lochbroom — Mathesons in Lochalsh ; Mackenzies in Kin- 
tail ; Gillanders or Rosses in Gairloch ; and Nicolsons in 
Lochbroom. Applecross was holy ground, and belonged 
to the Churih. 

In 142T Mackmaken or Matheson of Lochalsh is 
mentioned as leader of two thousand by Fordun ; he was 
vassal to the Earl of Ross, and concerned in his rebellious 
conduct. In 1449 Lochalsh was in the hands of Celestine, 
brother to John, Earl of Ross, granted to him by the Earl 
or his father, and confirmed by the Crown in 1464. What 
became of the Chief of Mathesons or how he was treated 
we do not know : at anyrate he had a new overlord. Sir 
Alexander of Lochalsh succeeded Celestine (died 1476), 
and in 1492 granted with other lands elsewhere to Ewin, 
son of Alan, Chief of the Camerons, the half of Lochalsh, 
14 merklands out of the 26, and the King confirmed it in 
1495, and renewed it in 1539. Meanwhile Sir Donald 
Gallda succeeded Alexander, and died in 1518, leaving 
his two sisters co-heiresses. One was married to Glen- 
garry, the other to Dingwall of Kildun. The other half 
of Lochalsh thus belonged to these two. They shared 
the farms between them. Thus the Davoch land of Bal- 
macarra belonged half to Glengarry and half to Dingwall, 
and so with the other holdings. In 1539 Glengarry has a 
direct Crown charter for his share of Lochalsh. Owing to 
circumstances arising from Blar-na-Leine and the conse- 
quent raid on Glens Moriston and Urquhart by Glengarry 
and Lochiel, their shares of Lochalsh passed by forfeiture 
into the hands of the Grants — Freuchy and Glenmoriston. 
But they don't seem to have reaped any benefit from this 


" sheepskin " allotment to them of these lands. Dingwall 
in 1554 sold his share of Lochalsh to Mackenzie of Kin- 
tail, who was fast acquiring territory and power ever since 
the fall of the Earldom of Ross. Freuchy in 1571 made 
peace with Glengarry, and over a marriage contract with 
Glengarry's son and Freuchy's daughter gave him back 
his lands and his (Freuchy's) share of Lochalsh. Lochalsh 
was thus in the hands of three men in 1571 — Glengarry, 
Kintail, and Grant of Glenmoriston (5 merklands). The 
quarrels between Kintail and Glengarry began about 
1580. Glengarry was not popular in Lochalsh — the old 
Mathesons had not yet, possibly, forgotten the Macdonald 
usurpation — and at anyrate Glengarry himself was unwise 
and harsh. The Mackenzie Chief fomented the quarrel, 
then openly intervened in it, hostilities breaking out 
about 1580, which went on intermittently till 1603, with 
much loss in blood and status to Glengarry. The final 
result was that in 1607 Glengarry was compelled to part 
with the lands of Lochalsh to his more astute rival of 
Kintail, who thus in that year acquired the whole of Loch- 
alsh save Glenmoriston's five merks and the Church lands, 
the latter of which, however, became his in 1610, and in 
1633 the Earl of Seaforth is retoured for all Lochalsh. 

The Mackenzies held Lochalsh for two hundred years, 
the last Lord Seaforth parting with it in 1801. The pur- 
chaser of Lochalsh was Mr, afterwards Sir (1818), Hugh 
Innes, who had made his fortune in London by commerce. 
He was some time member for the Wick Burghs, and, 
dying without issue, his property came into the hands of 
his grand-niece (Katherine Lindsay), v/ho married Mr 
Isaac Lillingston. The Lillingstons were much beloved 
by the Lochalsh people, and receive great praise from 
Duncan Matheson, one of the clan historians of the time. 
After the death of her husband Mrs Lillingston sold 
Lochalsh to Mr, afterwards Sir, Alexander Matheson in 
1851, whose son, Sir Kenneth, is the present proprietor. 


The place-names of Lochalsh are nearly all Gaelic. 
The Norse names are remarkably few — nine or ten in all. 
There is at least one Pictish name — Lundie and Loch 
Lundie (Lunde, 1495; Lundy, 1527). It is a common 
name in Pictland, from Fife to Sutherland inclusive. It 
is possibly the same name as we have in London, ancient 
Londinium. The root may be a nasahsed form of LUD, 
marshy, boggy ; Gaelic, LODAN. 

The Norse names are as follows : — 

Stromeferry. This is a hybrid ; " ferry " is Eng- 
lish and " strome " is the Norse " straumr," current, 
stream, applied to the Strome channel here. It is common 
in the Orkneys and in the Norse regions generally. 

Ulhava, an islet near Duncraig. This is the same 
as Ulva, near Mull ; it means " Wolf's Isle "— ULF-EY. 
Ulf may have been a person's name. 

DuiRINISH (Durris, 1548, Durness, 1554, Dowrnes, 
Durinische, 1607). This means Deer's Ness or headland 
— " dyra-nes." The name appears as Duirinish in Skye 
and Durness in Sutherland. 

Erbusaig (Arbesak, 1554, Erbissok, 1G33, pro- 
nounced now Earbasaig with two r's for euphony's sake). 
It appears to mean Erp's Wick or bay, Erp bemg a per- 
sonal name borrowed by the Norse from the Picts. The 
Gaehc form of Erp is Ere, a common name in ancient 
times, as in the case of Fergus Mac Ere, first King of 

Pladaig. The Norse chief root here is FLAT, flat; 
compare Pladda and Fladda for Flat-ey, " Flat Island." 
Whether AIG is a Gaelic termination or the Norse VIK, a 
wick or bay, is hard to say. 


SCALPAIDH, pronounced Scalpa, means Shallop or 
Ship River— " Skalp-a." Scalpa, in Skye, is Ship Isle, 
.and in the Orkneys it is for Ship-isthmus (EIDH). 

Reraig (Rowrag, 1548, Rerek, 1554, Rerag, 1607, 
1633), Gaelic Reiraig, seems to be for REYRVIK, Reed Bay. 
There is another Reraig in Lochcarron. 

AVERNISH (Awernis 1459, Awnarnys 1527, Avarrynis 
1548, Evernische 1607, Averneis 1633) ; Gaelic A(bh)arnis. 
It is likely Norse " Afar-nes," Big or Bulkyness. 

Ceann-AN-OBA ; Gaelic, " Ceann an oib." Head of 
the ob or bay ; the word OB is in Norse HOP. It appears 
as Obbe in Harris, as Oban, and Ben Hope in Sutherland. 
Ob-an-Duine is a little north of Plockton. 

These are the undoubted Norse names. It is tempt- 
ing to refer Loch Calavie, far eastward among the hills, 
to the Norse KALFR, calf, especialy as in the next Glen is 
Loch an Laoigh, beside which is Coire Seasgach (corry of 
the heifers). The difficulty in regarding it as Norse is 
two-fold : it is far inland and the pronunciation is Cail- 
bhidh, where the termination in " i " is unaccountable from 
Norse sources. There is a Glen Calvie in Strathcarron. 
Strathasgaig, Gaelic Srath-asgag, may be hybrid, Asgaig 
coming from the Norse " Aska-vik," Ash or Ship Wick ; 
but the Norse ASK has the vowel short, and this makes the 
etymology doubtful. No Gaelic root in ASG, FASG, or 
even TASG or SASG can be suggested. 

We shall now deal with the Gaelic names mentioned 
in the old records, commencing at the north end of the 
parish and working around the coast. 


Ardnarff, G. Ard-arbha, Ard-an-arbha ; " Height of 
the corn(land)." In 1554 Ardnanarf 15T4 Ardenarra, 1607 

Inchnairn ; G. Innis-an-fhearna ; " Inches or the 
Links of the Alder." In 1548, 1554, and 1607 Inchenarne, 
1574 Inchnairnie. 

Fernaig ; G. Fearnaig, " Placs of Alders." In 1495 
Fairnmoir, Fayrineagveg (there were two Fernaigs), in 
1527 Fayrnagmore, Fayrinaegveg, &:c. 

ACHMORE, "Big Field"; in 1495 and 1527 Achmoir, 
in 1548 Auchmoir. Along with it went Killochir (1548, 
1607) or Cuyloir (1527), a name seemingly lost. 

ACHACHONLEICH, "Field of the Straw or Stubble" ; in 
1495 Achechoynleith, 1527 Achchonelyth. 

Braeintra ; G. Braigh' an t-Srath, " Upper part of 
the Strath." In 1495 Brayemtraye, 1633 Breaintread, 
1548 Brayeyntrahe. 

Craig, Duncraig, from G. CREAG, hill. In 1548 
Cragy et Harsa, 1554 LIE Craig ; in 1607 Craig et Harsa. 
The latter name is unknown to the present writer. 

ACHANDARRACH ; G. Achadh-nan-darach, "Field of 
the Oaks." In 1495 Achenadariache, 1527 Achendariach, 
1548 Auchnadarrach. 

ACHNAHIXICH ; G. Achadh-na h-inich ; 1548 Auchna- 
howgych, 1554 Auchnaheuych, 1574 Auchinnahynneych, 
1607 Auchnahinginche, 1633 Auchnahenginche. Duncan 
Matheson spelt it Acha na Shinich, and he says that at 
Achadh-da-temaidh (Field of two descents) there the 
Mathesons used to rally as to a rendezvous when they took 


the field. They drank of the sacred stream of Altan- 
rabhraidh (Burn of the Murmuring) and started. If 
AONACH, fair, gathering, were dialectically feminine in old 
Lochalsh, as well it might, for the word was originally 
neuter, then we might explain the name as Achadh na 
h-Aonaich, "Field of the Fair" — even "Field of the Ren- 
dezvous." A word INGNEACH suits the old forms and 
pronunciation best. 

Balmacarra, G. Bail' mac-ara. The old forms are 
— 1548 Ballimaccroy ; 1554, 1607, and 1633 Ballamaccarra ; 
1574 Ballemakcarra(ne). The name looks as if it meant 
"Township of the M'Ara family" (M'Ara being genitive 
plural). The surname M'Ara or M'Carra is and was 
common in Perthshire ; but it is not found elsewhere. 
Balmacarra may be a corruption, like Ben Mac Dui ; per- 
haps. Whale-ton (Muc-mhara) ? 

AUCHTERTYRE, G. Uachdariridh, for Uachdar-thire, 
"Upper part of the Land." Old forms are— Wochterory 
(1495), Ochtertere (1527), Ochbertirie (1548). 

ACHTAYTORALAN, G. Achadh-da-toralan. Old forms 
are — Auchtatorlyne (1548), Auchtatorlane (1554), Auchrid- 
tidorillane (1574), Auchtatorrelan (1607). The word 
TORALAN or TORRALAN is of doubtful force ; it may be a 
derivative like TORRAN, knoll, from TORR, but the name 
Achadh-da-tearnaidh, already referred to, makes it pos- 
sible that here we have a name of similar form and force, 
viz., TORLUINN, descent, better TUIRLING. With Achtay- 
toralan went Ardach (1548), Ardache (1607), Ardacht 
(1574) ; It means " Highfield." 

NOSTIE. G. Nosdaidh ; in 1548 and 1574 Nosti, 1554 
Noyste, 1607 and 1633 Nostie. It seems to be from OSD- 
THIGH, inn, with the article in the dative (or locative) before 


it, as 'N-Osd-thigh, just as we have it in Nonach further 

Ardelve, G. Aird-eilghidh, " Height of the Fallow 
Land." In 1548 Ardelly, 1554 Ardelf (which suggests a 
locative Ard-sheilbh, " High-property,") Ardillie in 1574, 
1607 Ardelleive, 1633 Ardelve. 

CONCHRA, G. Conachra, which appears to be a deri- 
vative of CRO or CRA, an enclosure, fold (compare the Cro 
of Kintail, Cra in the north of Arran, and elsewhere in 
both forms). The CON is the old preposition CO, CON, 
with ; it means here a collection of CRO-s, the whole word 
Conchra meaning " Place of folds." In 1548 Connachry, 
1554 Concry, Conchra in 1574 and 1633, in 1607 Conchara. 

Sallachy, G. Salachaidh, "Place of Willows;" old 
Gaelic SAILECH (gen.), willow, now SEILEACH, the Scotch 
SAUGH, for SALCH, old English SALH. Sallachy is com- 
mon as a place-name. Compare Sauchie-burn for older 
Salchie (Stirling), where possibly the word is Scotch. Old 
forms are Sallach in 1548, Salche in 1554, Sallachie in 
1574 and 1633. 

Old names that seem to have dropped out of the run- 
ning are these — With Fernaigbeg go Fadamine (1495), 
Fynimain (1527), Fineman (1548), and Acheache (1495) 
Acheachy (1527), Auchcroy (1548, 1607). The two merk- 
lands of Culthnok, Achnacloich, Blaregarwe and Acheae 
are mentioned in 1495 and 1527 (Achiae in 1527) and 
later. With Auchtertyre appears Achich in 1548, Achiche 
in 1607. 

Names that do not appear in the old documents will 
now be taken. 


PORT-A-CHULLIN, G. Port a' Chuilinn, " Holly Port." 

Plockton, G. Am Ploc, " The Lump," applied to the 
humpy promontory which ends in Ruemore (G. Rudha- 
mor, "Big Cape"). Duaird is G. Dubh-aird," "Black 
Point. Lon-buidhe is " Yellow mead." 

Strathie, G. Srathaidh, " Straths." A plural loca- 

Sean-Chreag, "Old Rock." 

PORT-EORNA, G Port an Eorna, " Barley Port." 

Drumbuie, " Yellow Ridge." 

PORTNACLOICHE, " Port of the Stone." 

Palascaig, Loch Palascaig; doubtful. 

BADICAUL, G. Bada-call, " Hazel-Clump." There is a 
Badcall in Rosskeen and another in Eddrachilles. 

Kyle of Lochalsh. Kyle in G. is Gaol, a narrow, 
COILLEMORE, " Great Wood." 

Glen UdalAN. In Gaelic, UDALAN signifies a swivel 
or swingle tree. It is difficult to explain the connection 
here. It is likely that the river was first named Udalan. 
There is a Ben Udlaman on the confines of Badenoch. 

KiRKTON. The Gaelic is Clachan, "village," 
" church." The burying-ground is called Cnoc nan Ain- 
geal, " Angels' Knoll." 


KiNNAMOIXE, G. Ceann-na-moine, " Moss-head." 

ElLEAN TlORAM, " Dry Island " (a common name), 
is at the entrance to Loch Long, " Ship Loch," a name 
found in South Argyll and elsewhere. Camas Longart, 
" Bay of the Encampment," from LOXGPHORT. The River 
Ling is in Gaelic Abhainn Lumge, " Ship's River," con- 
nected with Loch Long. 

AULTNASOU, G. Allt-nan-subh, " Berry Stream." It 
was called Aultnasou in 1721. 

NONACH, G. Nonach. Loch na h-onaich, not far off, 
shows that we have here the article AN with OXACH. We 
may compare Onich, near Ballachulish, which is derived 
from Omhanach (locative Omhanaich), " Place of Foam." 

POLL-AX-TARIE, G. Poll-an-tairbh, " Bull's Pool," 
where the legendary battle between the Mathesons and 
Sutherland men took place. 

Patt, on Loch ]\Ionar, G. Pait, "Hump." The 
shootings of Riochan (Riabhachan, " Brindled Place,") 
and Sail-riabhach (from a hill so called, " Brindled Heel,") 
are near here. 

The names of the rivers, lochs and hills not already 
mentioned are easy, save in the case of Loch Monar. The 
word Monar seems to be merely a more phonetic form of 
G. MOXMHUR, a murmuring noise ; purling of a stream or 
of water would be its meaning in this case. Coire-na- 
sorna, near Loch Calavie, is interesting as giving a fem- 
inine genitive to G. SORN, furnace, gully; but the word 
was both mas. and fem. in early Irish. We have the cor- 
rect genitive in Loch Houm, which stands for Loch 


Shuirn, *:he loch taking its name from SORN. This is 
proved by the Dean of Lismore's line — 

" Leggit derri di wurn 
eddir selli is sowyrrni " 

(" An end of merriment between Shiel and Hourn ;" that 
is, in the Clan Ranald country). 






The name Hebrides is, like the name lona, due 
to a clerical blunder; and Hector Boece is the 
author of it. He misread the Classical Hebudre 
with a middle vowel u, as Hebrida, with a 
medial ri. The oldest form of the name appears 
in Mela (1st century a.d.) as H£emoda3, of which 
he says there were seven. Phny accepts Mela's 
Haemodas, and adds 80 Haebudes, while 
Ptolemy (2nd century) has only the Aiboudai, 
that is, Aebudse, 5 in number. These he 
separately names Aebuda, one and two, Ricma, 
Malaeos, and Epidium. An attempt has been 
made to identify the two Uists with the two 
Aebudae. Uist appears as Ivist in the old Norse 
poetry, and it has been possibly modified to 


sound like the Norse i vist, a habitation. In any 
case, we cannot be certain that either in Aebudae 
or in Uist we have anything resembhng a Celtic 
spelhng of the original word meant. Uist, as 
we now have it, has been handed down on Norse 
hps. Of Ptolemy's other three islands, Malaeos 
is clearly Mull; Adamnan mentions it also as 
Malea and the Norse as Myl. Adamnan besides 
mentions lona, that is, Hii or I, St Columba's 
Isle, and he further notices Coloso (Colonsay), 
Egea (Eigg), Ilea (Islay), Longa (Luing dr 
Lunga?), Sainea (Shuna), Scia (Skye), Terra 
Ethica (Tiree), and the unidentified ones, 
Airthrago, Elena, Hinba, Oidecha (Texa?), and 
Ommon. Many of these w^ere re-named by the 
Norsemen, and their locality can only be 
guessed. Ptolemy doubtless also means Skye 
by his Sketis, though it is placed eastw^ard of 
Cape Wrath and his Orcades Isles. 

That the names of the western islands before 
the advent of the Norsemen were Celtic is 
probable; that Celts inhabited them is equally 
so. The names Orcades is distinctly Celtic; 
the root is ore, pig, allied to the Latin porcus, 
and the English farrov:, for the Celtic languages 
have lost initial p in every native word. The 
name Skye, Norse Skidh, Adamnan' s Scia and 
Ptolemy's Sketis, has been properly identified 
with Gaelic sqiath or sgiadh, wing. The name 
of Malaeos, now Muile, that is. Mull, may come 


from a root mal, which Dr Whitley Stokes 
compares with Albanian mal, mountain range, 
border, Lettic mala, border; to which we may 
add Gaelic mala, eyebrow. The idea would 
therefore be " the mountainous island "^ 
" Muile nam morbheann." Ptolemy's Epidion 
has not been identified; but the root is clearly 
the British or Pictish epo, horse, and the Epidii 
of Kintvre must have been so named, as the 
Echaidhs and Eachanns of Gaelic old and new, 
from their horsemanship. The name Colosus 
in Adamnan, now Colonsay, Gaelic coll-asa, may 
have something to do with coll, hazel. The 
word Hii, or I, or lona, is extremely puzzling — 
Dr Whitley Stokes suggests a connection with 
Latin plus, holy, or Celtic i-ios. Tiree is in 
Adamnan Insula or Terra Ethica, " land of 
corn," eth being his form of old Gaelic ith, 

We are, so far, justified in assuming that the 
Western Isles were under the sway of the Celts, 
and that their inhabitants spoke a Celtic lan- 
guage until the advent of the Norsemen. The 
latter people appear first about the year 794, 
and terrible was the confusion and havoc that 
they caused. They completely colonised the 
Orkneys and Shetland, which the Celtic popu- 
lation never recovered; and almost the only 
remembrance of them there is the name of the 
Orkneys. Nowadays, the old idea that the 


Teutonic invaders of the Celts annihilated the 
previous population has been abandoned, both 
for Saxon and for Norse conquests. But if it 
was anything like the truth anywhere, we may 
claim it as such for the Long Island. The Gaelic 
names were clean swept out of the island; the 
present Gaelic names are post-Norse imports. 
But the further we go south among the isles, the 
less sweeping does the clearance seem to have 
been. Lews was evidently re-named by the 
Norsemen, but in the southern isles the pre- 
ponderance of Gaelic names shows that the 
Gaels were absorbed gradually, not extirpated, 
as they were in Lews. At present, the propor- 
tion of Norse names to Gaelic ones in Lewis is 
as 4 to 1 ; in Islay it is as 1 to 2 ; in Arran as 1 
to 8; and in Man it is about the same, or rather 
1 to 7 J. It is not clear if Gaelic was ever com- 
pletely exterminated in Arran and the more 
southerly isles. It is most likely that it was not; 
inter-communication between Ireland and Scot- 
land would help its continuance. The battle of 
Clontarf (1014) undoubtedly wrought harm to 
the Norse and Danish power. The Islands 
asserted their independence of Norway, but 
were cruelly subjugated again by Magnus Bare- 
legs in 1098. In 1156 those islands south of 
Ardnamurchan were ceded to Somerled, a Norse 
Celt, who started seemingly a Gaelic-Norse 
kingdom of the Isles. In 1266 the Hebrides 


were finally ceded to Scotland, after being 470 
years under Norse sway. The Gaelic language 
seems to have rapidly spread itself through the 
Isles in the time of the Norse decadence, and at 
the cession of the Isles. At the present time, 
to parody the expression " Hiberniores Hibernis 
ipsis," we may say that the Norsemen of Lewis 
at present are Goideliores Goidelis ipsis — more 
Gaelic than the Gaels themselves. 

We may quote a passage from the lays of 
Magnus Barefoot' s time, recounting his domgi 
in the Isles. It will also serve the purpose of 
showing the early twelfth century form of the 
island names: — "Fire played fiercely to the 
heavens over Liodhus (Lewis); he (Magnus) 
went over Ivist with flame: the veomen lo«t 
life and goods. He harried Skidh (Skye) 
and Tyrvist (Tiree) .... the Mylsk (people of 
Mull) ran for fear. Far over the flats of Sandey 
he warred. There was a smoke over II (Islay); 
the king's men fed the flame. Further south, 
men in Cantyre — Santiris — bowed beneath the 
sword-edge. He made the Manxmen — Man- 
verja — to fall." 

So much did the Gaelic population of the 
mainland feel that the Isles were Norse that the 
proper Gaelic name of the Hebrides has been 
Innse Gall, " The Islands of the Galls or 
Strangers." The word Gall now means a Low- 
lander or English-speaking person; originally 


the word was used by the Gael, as by the 
Eomans, to designate the Gauls of France and 
Britain, who came as strangers in contact with 
the Gael of Ireland in the earlier centuries of our 
era. Similarly the name Welsh comes originally 
from that of the Volcae, a people on the northern 
border of Gaul, and marching with the Teutons, 
who named all Celts after them. 

Dean Munro in his survey of the Isles in 
1549 mentions and describes 209 of them. Of 
these names many are repeated — there are about 
eight Fladdas, and many more are easily under- 
stood on the score of derivation; so that at this 
part of our subject we need only refer to the 
more important islands. The first name that 
claims attention is that of Lewis. This in Norse 
times was Ljodh-hiis, the contemporary Gaelic 
of which was Leodhiis. The Norse word may 
mean "loud house" and "lay house," the 
latter meaning either song or people; but these 
meanings are unsatisfactory, and resort has been 
had to Gaehc, old and new. Martin (1703) 
derives the name from a Gaelic word leog, 
" water lying on the surface of the ground," a 
word for which he is the first and sole authority. 
The fact is that we do not know the pronuncia- 
tion of Martin's word. The modern Gaelic 
name is spelt Leodhas, and pronounced in Gaelic 
phonetics as Leo's. In the present state of our 


knowledge, it is impossible to say what the name 
may mean. 

The Harris portion of the Long Island is in 
Gaelic called Na h-Earradh. In old documents 
of three centuries ago the name appears vari- 
ously as IIar(r)ay, Her(r)e, Herrie, and, with a 
plural form, Harreis (1588). Captain Thomas 
adduces other places so called, such as Harris in 
Bum, Herries in Dumfries, Harray (Orkney), 
and Harrastadhir in Iceland; and he further 
derives the root word from Norse har, high, with 
a plural havir, and a comparative h^erri, higher, 
allied to a noun haedh, height, whose plural is 
haedhir, which, he thinks, might develop into 
har-ri by a shifting of the r. The name Harra- 
stadhir is rather to be referred to Norse liarri, 
master, king, and the Orkney Harray may be of 
the same origin or descended from the word 
herad, district. The meaning which Captain 
Thomas attached to the name Harris was that 
of " The Heights." Unsatisfactory as the 
phonetics are, this is the best derivation as yet 
offered for Harris. The final .s in the English 
name Harris is, of course, the sign of the plural. 

St Kilda is known in Gaelic as Hirt, which is 
its old name, the term St Kilda being only two 
centuries old and of doubtful origin. In ancient 
Gaelic irt signifies " death," and possibly the 
island received its name from its remote western 


position, for the Celts connected the West with 
the abode of the dead. 

The Uists we have akeady discussed; and 
before passing on we may notice the most im- 
portant islets in the Sound of Harris and there- 
about. Taransay is St Taran's ey or island; 
Scalpay (in Harris and Skye) is " Ship-isle," 
from Norse Skalpr, a ship, whence also the old 
Orkney Skalp-eidh or ' ' ship isthmus ' ' ; Scarp is 
the Norse Skarpr, sharp, and is the same as the 
English scarp, escarpment; Pabbay is a common 
name, appearing in Iceland and the Orkney and 
Shetland Isles as Papey or Papa, and it signifies 
Pope's or Priest's Isle; Shillay, or Shelley 
signifies "seal-isle"; Ensay is for engis-ey, 
" meadow-isle "; Tahay is for Ha-ey, high-isle, 
the same as the Hoy of the Orkneys; Coilsay is 
for " gill's isle," gill being a ravine; Boreray is 
for Borgar-ey, that is, "Burgh-isle" or 
" fortress-isle," not, as Captain Thomas sug- 
gested, Boru-ey, " bore-isle " — thus, Dun- 
vorrerick is the Dun or Fort of Borgar vik, that 
is, Fort-bay. Personal names appear often in 
these island names — -Bernera is for Bjorn's-isle 
(Bear-isle); Grimisay is for Grimm's-isle; Her- 
metray is for Hermund's isle, as Gometra, 
further south, is for Godmund's isle; Eriskay is 
doubtless, as Professor Munch said, Eric's isle, 
by metathesis of the s; and Barra is St Barr's 
isle. The Flannen Islands (na h-Eileanan 


Fiannach) are called after St Flannen. Taransay 
we have already explained as St Taran's isle. 

The name Benbecula is a curious perversion 
of the Gaelic Beinn-a'-bh-faoghla," "Hill of 
the Ford.'' Scaravay and Scarba are for skarf- 
ey, "cormorant (skarfr) isle"; Haskir is the 
Ha-sker, high-skerry, and Heisker is probably 
for heidh-skar , bright-rock ; Vallay is for ' ' field- 
isle," from voUr, field; Lingay is for ling or 
heath-isle, from Norse lyng, heath; Hellisay is 
from hellir, cave, also appearing in Orkney. 
The names Vatersay and Sandray easily explain 
themselves, from water and sand (vatn and 
sandr of Norse); Foula, or Fula, is for "fowl 
(Norse fugl) isle." There are four islands called 
Wiay, Dean Munro's (1549) Buya or Bywa, now 
pronounced as Fuidhaidh, to which we may 
compare the Icelandic Veey or "house-isle," 
Ve being very common in Norse place-names. 
Names like Fuday, Fiaray, Killegray, and Min- 
guiay, (pronounced MeaWa), we pass over as 
inexplicable to us, and come to Skye and its 

Eaasay and Eona are slumped together in 
the Norse name appHed to them, viz., 
Eauneyjar, the raun-isles, a word which Pro- 
fessor Munch explains as meaning experiment. 
Despite the absence of initial h, we should refer 
it rather to hraun, lava, rough ground, which is 
characteristic of the several Eona Islands that 


exist. The Norse Eaunen are bare rocks in the 
!5ea. Captain Thomas explained Raasay as " roe 
isle," from ra, roe; but this would give a geni- 
tive rar, not ras. It is difficult to say what the 
root word in Raasay is. Hrauns-ey is possible. 
There are some eight Fladda Isles on the Weat 
Coast; these all mean Flat Isle, the Norse 
Flat-ey. Another common name is Soa or 
Sheep-Isle, from the Norse Saudhr, sheep; 
again Haversay is for hafrs-ey, " he-goat isle." 
Isay is ice-isle; and Trodday appears to mean 
pasture isle, from trodh, pasture, a name which 
appears in Trotternish, which Captain Thomas 
strangely refers to trylldr, enchanted, or Troll- 
ness, though how this result could phonetically 
occur one cannot see. 

We now come to the parish of Small Isles, 
which comprises Canna, Rum, Eigg, and Muck. 
The first name — Canna — has been explained 
from the Norse Kanna to mean the ' ' can shaped 
isle"; it appears as Kannay in 1549. Rum, 
pronounced like the English " room," appears 
in the Annals of Ulster under the year 676 with 
the genitive Ruimm; and Dr Stokes refers this 
" lozenge-shaped island " to a root identical 
with Greek rhombus. As the root of this 
appears to be vreng, wrench, the comparison is 
doubtful. Nor is Captain Thomas's derivation 
at all happy ; first he refers Uist to I-fheirste, or 
Isle of Fords or sandbanks (Gaelic fearsad), and 


then Rum is taken from I-dhmim, ridge-isle . The 
supposed Gaehc i for isle is, as Dr Stokes says, 
clearly the Norse word ey borrov/ed ; as a matter 
of fact, it does not exist in Gaelic at all. Eigg 
is Adamnan's Egea, and may be referred to an 
oblique form of Gaelic eag, a notch. Muck is 
good Gaehc; it is " pigs' isle " ; Eilean nam muc. 
There is also a Horse Isle near. 

Further south we meet with Coll, Tiree, and 
Mull, names already discussed. The name 
Gunna recalls the Norse gunnr, war, so common 
in proper names among the Norse. Lismore is 
Gaelic — the great lios or enclosure. The islands 
variously called Luing and Lunga or Lungay, 
seem all to be of Gaelic origin; and to contain 
the Gaelic word for ship (long) as base. Lunga 
and Lungay have certainly submitted to Norse 
influence. The two isles called Shiuna, and the 
isle of Shona, seem to be formed from Norse 
sjon, sight, a root which appears in Norse place- 
names in connection with " scouting" positions. 
Kerrera is in the Sagas called Kjarbarey. and 
possibly means " copse isle." The Calf of Mull 
is in Norse Mylarkalfr, Mull's Calf, and in Gaelic 
it is Calbh. These Calf islands are common. 

Passing island names like Treshinish, Erraid, 
and Seil, we meet with Ulva, that is lilf-ey, or 
wolf's isle, Ulf, or Ulfr, being really a person's 
name here ; Staffa is staff-ey, from stafr, a staff, 
referring to its basaltic pillars, as Professor 


Munch pointed out; Jura is, as the same 
authority says, for Dyrey, that is, Deer's Isle, 
whence also Duirinish and Durness, Deer's Ness. 
Colonsay (Norse Koln) is in Adamnan, as we 
already sa^Y. There are some half-dozen isles 
called Oronsay, Gaelic Or'asa, without n, and 
Captain Thomas happily explained the name as 
Orfris-ey, from the Norse orfiri, ebbing. As 
Vigfusson says, Norse " Orfiris-ey is the proper- 
name for islands which at low-water are joined 
to the mainland." And this is true of the 
Scottish isles so called. Oigha, of which there 
are two at least, is pronounced Gidhaidh, and 
appears in the Sagas as Gudhey or Gudey, that 
is, God-isle, or good-isle. Arran is in the 
Norse called Hersey or Herey, doubtless an 
attempt at the Gaelic name Arann. This is an 
oblique use of Gaelic ara, a kidney, as Or 
Cameron pointed out. The Arann isles off 
Galway are similarly named and explamed. 
The island is kidney-shaped. Bute is in Norse 
Saga and modern Gaelic called Bot, but what 
the name means it is not easy to say. The 
Cumbraes were called by the Norse Kumreyjar, 
the isles of Kumr, which name is usually ex- 
plained as referring to the Kymry or Welsh of 



The ubiquity of the Norsemen in the Western 
Isles is indicated further by the fact that even 
lonely St Kilda fell under their sway, and its 
leading features show Norse names. The names 
of the adjacent islets are Norse; Boreray and 
Soa are the isles of the Borg (burgh), and the 
Saudlir (sheep). A hill on the east of St Kilda 
itself is variously given as Oiseval or Ostrivail, 
which stands for Oserveaul — that is, Norse 
Austr-fell, or "East-hill." The Norse fell, 
hill, as a rule, appears in Gaelic as bhal or val. 
Other Norse words undergo wonderful changes 
in passing into Gaelic, so much so that at times 
their own mother (language) would have diffi- 
culty in recognising them. Norse setr, a seat, 
holding, appears in Gaelic as siadair or seadair, 
and in the place-names as shader. Bolstadhr 
and Bustadhr come to be bos, bost, or bus ; and 
fjordhr, genitive fjardhar, a firth, becomes 
terminally art, ard, ord, ort; and it may even 
disport itself as port, Snizort is Sneisfjordhr, 
spit-frith; Cnoideart is for Cnut's or Canute 'f 
Frith, etc. Violent initial changes also take 
place in borrowing these Norse words. Many 
Norse names begin with h, and it is a peculiarity 
of Gaelic that /i, as Macalpin humorously re- 
marked in regard to the singularity of Highland 


character and institutions, though not recog 
nised as a letter in GaeHc, " is used not only in 
every word, but almost in every syllable 
expressed or understood." It is a parasitic 
letter, and leans upon some other consonant. 
Hence Norse words beginning with h may be 
supported by a t. Norse holmr appears as tolm, 
genitive tuilm, an island or inch. According to 
Captain Thomas, this word appears terminally 
as am, um, while in the Northern Hebrides it 
becomes by metathesis " mol," as in Kisa-mol, 
for Kastel-mol. Compare Cobhsamul or Cos- 
mul, Linmul, etc. Habost, that is, ha-bolstadhr 
or high-town, appears as Tabost ; Loch Thamna- 
bhaidh stands for t-hamna-vagr, that is, haven- 
voe, haven-bay; Loch Thealasbhaidh is for 
Hellis-vagr, cave-voe, etc. 

In considering the place-names of the 
Western Isles, we intend to utilise more or less 
in full the work of the late Captain Thomas, 
R.N., W'ho wrote one or two papers on this sub- 
ject for the Society of Antiquaries, in whose 
Transactions they lie buried unknown to the 
general public. The first paper on the " Extir- 
pation of the Celts in the Hebrides " appeared 
in 1876, and the other on " Islay Place Names " 
in 1882, wherein Captain Thomas was helped 
by the well-known Gaelic scholar, Mr Hector 
Maclean, Islay. We shall take in alphabetical 


order the leading Norse words that enter into 
the composition of these Island place-names. 

A, that is, d, a river. This word, which 
forms the stem of so many place-names in Ice- 
land, is rare in Lewis and the Isles. Laxa in 
Iceland, and Laxa of Shetland, are synonymous 
with Laxay of Lochs in Lewis. It stands for 
lax-a, that is, salmon-river, lax being the Norse 
for salmon. Other salmon rivers will be men- 
tioned when we come to dalr, a dale. Few of 
the small rivers in Lewis have distinctive names, 
but the Creed seems to be an exception, and tells 
of odoriferous plants and flowers; for krydd 
means spice, and krydd-jurt signifies spice- 

Ass, that is, ass, a rocky ridge. A strange 
corruption has befallen this word in Shetland, 
where Vind-ass has become Wind-house. So 
Burn-house gives us the well-known name of 
Burns. There are two places called Valtos in 
Lewis, and a third in Skye, and there is no oss, 
that is, oyce or river-mouth, there. Furthe]', 
we have Garry Valtos in South Uist. In the 
Orkneys, Waldbrek, Waldgarth, and South Wald 
seem to be cognate words. Assuming this, we 
may put Vold-ass, that is, field-ridge, down as 
the original form where Void is an olden form 
of Vollr, a field. For the phonetics we may 
appeal to Norse threskjoldr, a thrashing floor, 



where we have voUr as oldr; and further, that 
the T.awmaii in Slietland in 1307 dates from 
Tingvold, that is, thing-voUr, the field of 
assembly, which appears in Dingwall also. 
Perhaps Valdaras in Iceland may be exactly our 

Bekkr, a rivulet, brook. This rare word in 
Norse topography appears to occur only in Die- 
bek in Harris, the Gaelic of which is Ceann- 
Dliibig. Dubec in Skye is no doubt the same 
name. As Captain Thomas remarks, we do not 
know the meaning of the prefix, except that it is 
not Duhh, Gaelic Mack. 

Bakki, bank. It was hardly to have been 
expected that this word should have been retained 
so near its original form as in Back, Stornoway; 
Hdboc — Highbank (Gaelic form Tabac), in 
Bernera of Lewis ; Bakka in Taransay (Taransay- 
banks), Harris; and Baclid in Barra. Bacca- 
skill occurs in the Orkneys; Backa and Bacca in 
Shetland; while Bakki is the name of thirty 
farms in Iceland. 

Bol-stadhr, a homestead. This word is 
widely diffused over the northern and western 
islands. In " Landnc4mab6k " it only occurs 
twice, and it forms no compound except with 
Breidh — broad, which is repeated eight times 
(Breidhabolstadhr). In the Shetland directory 
there are (2) Busta, (4) Bousta, but in combina- 


tion it is written (27) -bister, and but once 
-buster only. In the Scottish dialect i has fre- 
quently the sound of u. The Orkney rental of 
1595 contains (44) -bustar and (3) -busier; only 
a few of which are named after men. In the 
Lewis rental Bolstadhr occurs as Bosta in 
Bernera, Uig; and when used as a generic term 
it is shortened to -bost. Many of these names 
are easily interpreted; thus, Melbost — there are 
two of them — is for Mel-bolstadr, Links-Farm; 
Leurbost, Leir-bolstadr, mud or clay farm: 
Ci'ossbost, Kross-bolstadr, Cross-Fajm; Calbost 
is shortened in the same way as in the Orkneys 
and Shetland (Caldale, Calback) from Oaldbost, 
Kald - bolstadhr, Cold -Farm. Garbost, as 
written by Martin, would be pronounced Garta- 
bost by the Gael, and was originally Geira- 
bolstadhr, Geirr's-Farm; Geirr is a proper 
name. The Orkneys give Garraquoy; Shetland, 
Garragarth or Gerragarth; and Iceland, Geira- 
bolstadhr. It is but right to say that Geira may 
mean a " gore " or sHce of land. 

At Shawbost, on the west of Lewis — vari- 
ously written Sheabost, Shabost — is a lake, into 
which the sea sometimes flows; this is the Sjar. 
Loch Seaforth gets its name from the pent up 
salt lake, Saer, which forms its head; hence 
Scefjoidh, Ssefirth, Seaforth; and the oyce at 
Kirkwall is called the " Little Sea " ; Shawbost, 
then, is Sja-bolstadhr, that is, sea-lake-farm. 


There are two Habosts in Lewis; neither of 
them are upon high ground; the adjective, there- 
fore, is the same as in so many places called 
Holland, Hallandi, Hall-lendi, in the Orkneys, 
and the still more common Houlland in Shet- 
land. Hahost has been Hall-holstadr; from 
Hallr, a slope, dechvity. So Captain Thomas 

Swanibost is the same as Swanbustar, in the 
Orkneys; and is cognate with Swynasetter in 
Shetland, and Sveinseyri, and Sveinavatn in 
Iceland. Swanibost stands for Sveina-bolstadr, 
Svein's-Farm; from Sveinn, a proper name. 
Shelibost (the Gaelic form of which is Seilabost), 
in Harris, is identical with Skelbustar in the 
Orkneys, and cognate with Skeljavik in Iceland; 
and in its Icelandic form is Skeljar-bolstadhr, 
Shelly-Farm; from skel, a shell. Besides Nisa- 
bost, in Harris, there is another in Skye, and 
Nesbustar, in the Orkneys; all of which repre- 
sent Nes-bolstadhr, Ness-Farm, from Nes, ness. 
Horgibost, Harris, must be written in Gaelic 
Torgabost; the Norse horgr signifies a heathen 
place of worship. Captain Thomas thought 
that in the Orkneys it appears as Howbister ; and 
cognate names in Shetland are Houby, Huxter, 
Hogsetter. "' In this case," he adds, "as in 
several others, the name has suffered less 
change among the Gael than in the northern 
islands. On this farm is a fine Cromlech, figured 


in the Crania Britannica, hence its name, Hauga- 
bolstadr, How-Farm; from Haugr, How, cairn, 
sepulchral mound." As a matter of fact, horgr 
and haugr are different words of Hke meaning. 

Borg — 1, a small dome-shaped hill; 2, a wall, 
fortification, castle. In Iceland ten different 
places are called Borg, but " it may be ques- 
tioned whether those names are derived simply 
from the hill on which they stand (berg, bjarg), 
or whether such hills took their name from old 
fortifications built upon them; the latter is more 
likely, but no information is on record, and at 
present ' Borg ' only conveys the notion of a 
hill." In Shetland Borg is still represented by 
"Burgh" in two places, but the influence of 
Scottish speech has changed it to " Brough ' ' in 
eleven others ; in the Orkneys also ' ' Brough 
prevails. In the Hebrides, when written in 
Enghsh, it is " Borve " ; in Harris (twice); in 
Barra, and in Skye. Borve, in Barvas, Lewis, 
appears as Borg, Bora (error for Borva), 
Barove; and the Gaelic form is Borgh (in pro- 
nunciation the r is duplicated, Bor-rgh); hence 
arises the English form Borve. The name is, 
archasologically, of great importance, for in 
Shetland, the Orkneys, and the Hebrides it 
almost always indicates the location of a pre- 
Norse Pictish tower; but there are a few excep- 
tions, at anyrate in the Orkneys, viz., Burrow 


Head, Stroiizii; Bunvick, Saiidwick; and the 
Brough of Birsa. 

So completely is the original meaning of the 
word forgotten in the Hebrides, that it is usual 
to put Dun (that is, castle) before it; thus, Dun 
Borgh (grammatically Dun Bhuirgh) means 
Castle-castle. Captain Thomas thinks Professor 
Munch wrong in saying " that it (Shetland) has 
had no fixed settlers upon it before the arrival 
of the Norsemen." — (P. 90, Mem. Soc. Nor. 
Antqs., 1850-1860). He has here for a moment 
forgotten the Borgir or Pictish towers, which 
have never been claimed as Scandinavian, and 
are consequently pre-Norse. Says Captain 
Thomas: — "But my more immediate business 
is with a long paragraph on pp. 103-104 of the 
same volume, to this effect — Burra, in Shetland, 
following the analogy of the Orkneys, should be 
Borgarey, but it is almost certain that in ancient 
times it was called Barrey; this theory depend- 
ing on the statement that a part of King Hacon'e 
fleet, coming from Norway, passed south of 
Shetland, sailed ' vest fyrir Barreyjarfjordr,' 
and saw no land till they made Sule Skerry, west 
of the Orkneys. Barreyjarfjordr was, therefore, 
the Bay of Scalloway, and the present Burra was 
Barrey. To all this it is answered, that a ship 
leaving Norway and seeing no land until she 
arrived at Sule Skerry must have passed between 
Shetland and Fair Isle, which part of the sea 


may very well have been called Fiithareyjar- 
fjordr or Fridhareyjarfjordr, i.e., Fair Island 
firth, and that Barreyjarfjordr is certainly a mis- 
copy of either of these names, so that the ship 
would not enter or be near the Bay of Scalloway. 
Besides, I myself have been on the site of the 
Picts' Castle, of which the stones were carried 
away to build the pier of Scalloway. Burra in 
Shetland, like Burra in the Orkneys, is Borgarey, 

" It follows that Alfdis, Konalsdottir, of 
Barrey (Barreysku), did not come from Shet- 
land, and we are at liberty to suppose that she 
was a native of Barra, Hebrides. We are told 
in Gretti's Saga (Danish translation) that the 
father of Alfdis, the Barra girl, was Konall ; her 
grandfather, Steinmoor; her great-grandfather, 
Olver Barnakarl {i.e., Olver the child's-man); 
he obtained this honourable title because he 
objected to join in the Viking sport of throwing 
the children of their victims up into the air and 
catching them on the points of their spears. 

" There is no further mention of KonAll; it 
may be hoped he met an early death; Alfdis 
would then come into the family of Ofeigr 
Gretter, her uncle, who had fled, with all his 
family and servants, from Harold Fairhair to 
the Barra Isles. It may be gathered that these 
islands formed the stronghold of a clan of 
Vikings; and a cousin of Alfdis, Aldis by name, 


was there married lo a woudeu-legged Viking. 
Ultimately, we are told, they all went to Iceland; 
but the topographical names prove that either 
some remained behind or that other Vikings 
supplied their places — probably both. Alfdis, 
the Barra girl, was married in Iceland to a 
grandson of Olaf the White, King of Dubhn." 

Bragor, in Barvas of Lewis. This word does 
not occm" in Oleasby's " Dictionary," perhaps 
from the misfortune of neither author nor editor 
being nautical men. Braga is applied to reefs 
on which the sea breaks wdth extra violence, and 
Bragor is named from the shoal water lying sea- 
ward of it. Mackenzie's chart has " Bragd " 
for a reef oft' Skegersta, Lewis; and if this is not 
— but probably it is— a clerical error for Braga, 
it would show how well the old Norse forms have 
been retained in Lewis, for the word is probably 
Bragd-arr, and formed from " Bragd," the 
" fundamental notion of which is that of a 
sudden motion." In the Orkneys are two reefs 
called Braga, and Break-ness is Bragir-ness. 
(Captain Thomas). 

Briiy a bridge. This word is represented by 
Bru in Iceland; Brow, Brugarth, in Shetland; 
and Brogar in the Orkneys. Brue, in Lewis, is 
at the outlet of Loch Barvas. 

Beer, hoer, hyr — (1) a town, village; (2) a 
farm, landed estate. The only certain " by " 
in Lewis island is Eoropie, which has caused the 


ridiculous appearance of ' ' Europa Point ' ' on 
some maps, and explained as meaning the 
extremity of Europe. Eoropie is simply Eyrar- 
baer, i.e., Beach-village; from Eyrr, a beach. 
In Shetland they would say, " The boat is at the 
ayre " ; that is, on the beach as distinguished 
from rocks. There are at least four islands in 
the Outer Hebrides and two in Skye bearing the 
name Oransay, Ornsay. " In every case that 
I know of they are connected at low water by a 
reef to another island." (Captain Thomas). 
The real name is Ortiris-ey, or Ebb-isle, from 
orfiri, ebbing, not from eyrr, beach. 

Dalr, a dale. Dcel, a little dale. There are 
over 130 names compounded with dalr in the 
Landnamabok, and the " dales " are propor- 
tionally numerous in the Orkneys and Shetlands. 
They are scarcely so frequent as farm names in 
Lewis. Swordale occurs twice in Lewis, and 
the map of Skye has three dales of that name. 
The Ny Jardharbok has Sward boeli. Swordale 
is for Swardhardalr — Sward-dale; from Norse 
Svordhr, sward, green-turf. Swordale in Lochg 
contains a coppice of willows and birches which 
are the last living trees of the native forest, of 
which the name is commemorated in the 
adjacent " Birken Isles." Laxdale indicates 
the presence of a salmon river. There are no 
salmon caught in the Orkneys, nor is Lax con- 
tained in their name system. Neither are there 


salmon in Shetland, yet there are Laxlirths and 
Laxa, so that either the salmon have deserted 
the country or the Northmen have given the 
name of Lax to the fine sea trout. Laxdale, in 
Lewis, and Lacasdle, in Harris, are synonymous 
with Laxar-dalr in Iceland. Salmon-iiver-dale; 
from Lax, a salmon. Eoradale is written for 
Eyrar-dalr — Beach-dale; from Eyir, a beach. 
Hodel, spelled Rodle and Eoudil in the same 
rental, is cognate with Roeness, Shetland, and 
Raudhanes, Iceland, and must have been 
Raudhi-dalr — Red-dale; from Raudhr, red. 
Ranigdale, a wretched place on the shore of 
Loch Seaforth, is probably. Rannveigar-dalr ; 
from Rannveig, a proper name. On the west 
side of Lewis there are Dale-Beg — Little Dale, 
and Dalemore — Great Dale; and als(i Noith and 
South Dale. All these are sharp, little valleys, 
and their original has been Dael — a little dale. 
But it has to be remarked that when dale pre- 
cedes, the combination is of Gaelic origin, 
though the word dail is borrowed. 

Eidh, an isthmus, neck of land. In 1576 
Eidh in Shetland had become Ayth. now Aith, 
but a much greater change took place with 
Eidhs-vik, which in 1576 was Aythiswick, but 
now Ea-swick and E-swick. Eidh, in the Ork- 
neys, is often very corrupt. It remains almost 
intact in Aith, Aithstown ; less conspicuous in 


Ai-sdale ; but Haugs-eid has become Hoxa ; 
Eidh-ey, Eday; and Skalp-eidh, Scapa. 

Eidli has many strange forms in tJie 
Hebrides : le, Ey, Y, Ay, Eie, Huy, Ui, Vy.', 
Uiy, Uie, Eye; written in Gaehc it is Uidh 
(pron. Oo-ee). Uiy, Eiy, in Taransay, is simply 
Eidh — isthmus. Branahuie, Stornoway, is 
better written in the Gaehc form, Braigh ua 
h-Uidhe; where Uidhe represents Eidh, Aith- 
' isthmus. Braighe is the Gaehc for upper part, 
upper end; and Braigh na h-Uidhe means the 
farm at ' the upper (nearer) end of the isthmus ' ; 
Uie-head occurs again at Vattersay, Barra. The 
peninsula of Eye is near Stornoway. 

Endi, the end. Mossend, Stornoway, as it 
stands, is modern Enghsh, but it is hkely to have 
been derived from Mos-endi — Mossend; from 
Mosi, moorland, moss. Tobson; any Scandi- 
navian name beginning with H may, when used 
as a Gaelic word, have an initial T, and th(^ H 
deleted to put it in the nominative case; thus 
Tobson is a Gaehc from of Hops-endi — the end 
of the hope or tidal lake. 

Ey, an island. Ey, in some form (a, ay, in 
Gaelic aidh), is the termination of the name of 
nearly every island in the Hebrides that is 
smaller than a land, or larger than a holm. 
Captain Thomas notices only those that are 
named in the rentals. He says — " There are 
three islands, in the Outer Hebrides, called 


Bernera, for Bjarnar-ey; Bjorn's-isle, from 
Bjorn, a proper name. It is to be noticed, the 
names of these islands are pronounced by the 
people, not as they are written in English, but 
in their Norse forms (e.g.) Be-ornar-ay {vide 
' Princess of Thule '). Besides Scalpa, in 
Harris, there is another in Skye, and both have 
snug httle harbours; there is also Scalpa, Skalp- 
eidh, in the Orkneys. Scalpa is for Skalpey, 
Ship's-isle, from Skalpr, a kind of boat or ship — 
shallop. There are two Shellays, one belonging 
to Harris, the other to North Uist. Shellay is 
the Gaelic pronunciation of Sellay, and this is 
for Sel-ey, Seal's-isle, from Sel — seal. Ensay 
is a remarkably fertile island, and well deserves 
the name of Engis-ey, Meadow (Grassy)-isle ; 
from Engi, a meadow. Scarp is again repeated 
in Barra as Scarpamutt. Scarp, more properly 
Scarpay, is for Skarp-ey, Scarped or Cliffy-isle; 
from Scarpr, scarped. Hermitray is indeter- 
minate. [It is Hermundarey, or Hermund's- 
isle.] Taransay, St Taran's island. The ruins 
of his church are still traceable, and a stone 
cross from it is in the Museum. A curious 
legend is related by Martin (West. Isles, p. 48); 
but I suspect he has inverted the names, and 
written ' Tarran ' for ' Che ' (in later copies, 
' Keith '), and the contrary. There is a St 
Tarannan, Abbot of Bangor, commemorated on 
the 12th June. There are four islands having 


the name of Pab-ay, in the Outer Hebrides; 
another in Skye; two (Papa) in the Orkneys; 
three (Papa) in Shetland; and one (Papey) in 
Iceland. The name is very interesting, for it 
indicates that Culdees, Celi-de, Servi-Dei, were 
located there before the devastation by the 
Northmen. Pabay, Pabbay, for Pap-ey, 
Priest's-isle; from papi — priest." 

Fjordhr, a firth, bay. This word fjordhr 
takes many forms in the Hebrides— -such as 
"port, fort, forth, furt " ; and by aspiration 
becomes in Gaelic orthography fhurt; hence 
written phonetically " ort, ord, irt, urd," etc. 
In the Orkneys and Shetland it is " firth." 

" Resort on the west, and Erisort on the east 
side, divide the mountains from the lower 
(though anything but level) part of Lewis. I 
believe them to be the same word. On looking 
into the history of the word, it is found written 
' Eriford, Erisport, Iffurt (error for Isfurt), 
Herrish — Arisford ' with the Gaelic Loch pre- 
fixed. These words plainly represent Herris- 
firth." Such is Captain Thomas's idea We 
have already remarked on the name Snizort and 
Knoydart. Further, there are Gruinard (grein- 
fjordhr, the split-firth), Moidart (Modha, a large 
river, loam, vapour), Enard, Eport, Skiport 
(ship-firth), etc. 



Following the alphabetical order of the chief 
Norse words which enter into the topography of 
the Western Isles, we came, at the end of our 
last article, to the word fjordhr, or firth; and 
its appearance in the names of places, after six 
hundred years of Gaelic pronunciation, as ard, 
ort, and port, though extremely puzzling to an 
outsider, is just what we might expect from the 
laws that regulate aspiration and accentuation 
in Gaelic. It is the rarer words which precede 
these Gaelic terms that puzzle the student of 
place names. Thus Gruinard has been vari- 
ously explained as green-firth, split-firth, and 
shallow-firth (Norse, Grunna-fjordhr), and the 
latter is doubtless the correct derivation in the 
case of the Islay Gruinart, for shallowness does 
characterise the firth there. 

We left, in our last number, an important 
word out of our alphabetic list, in our eagerness 
to discuss its compound form, bolstadhr. This 
is the simple bol, which means a building or 
farm. Thus the Icelandic Kirkjubol appears as 
the Sutherlandshire Kirkiboll and the Tiree 
Kirkapool. The island of Tiree, indeed, shows 
a plethora of these bols, but in the deceptive 
form of poll or pool — there are Crossai)oll 


(Cross-town), Barrapoll (Barley-town), Hellipoll 
(Helgi, holy), and VassipoU. In Islay there is 
Corsopoll, and in Coll, Crossipoll (both Cross- 
town), and there is Harrapool in Strath, Skye, 
a name which must be connected with the Norse 
Harrastadhir, discussed in connection with 
Harris. Sutherlandshire presents, among 
others, Eriboll (Beach-town) and Skibo (Skelbol, 
Shell-town); and Ullapool shows the same word 
combined with a personal name (Ulli). 

Resuming our alphabetical list of the most 
general Norse names, and still relying largely 
on Captain Thomas, the first we take up is Fell, 
fjall, the Norse for hill. It appears in Gaelic 
terminally, as bhal, val, and even in the good 
Gaelic form of Mheall, which has the same 
meaning. Compare Coinn-mheall in Suther- 
landshire. It is the commonest name for a hill 
in the Outer Hebrides, and, as its force is now 
lost, the prefixing of the Gaelic Beinn (hill) 
duplicates the sense. "It seems at first un- 
accountable that the lakes and hills in the 
uninhabited parts of these islands should have 
retained their Scandinavian names to this day. 
But, in fact, the whole country was divided for 
a pasture among the town-lands of the coast, and 
about midsummer nearly all the people removed 
with their herds to the moors, so that the most 
desolate spots were yearly inhabited and de- 
pastured; and it is from this cause that so many 


of tlie place-names have been remembered. 
Copeval in Harris is for Kupu-fell, Bowl 
(6haped)-f ell ; from Kupadhr, bowl-shaped con- 
vex " — (Captain Thomas). Goatfell in Arran 
is unmistakable. 

Gardhr, an enclosure, garth. In the " Com- 
playntes " of Shetland, 1576, gardhr becomes 
garth or gar, seldom goird, gord, gorde. In 
the "Old Rental" of the Orkneys (1503), 
gardhr is represented by garth, which in 1593 
has generally become gair, and is now commonly 
ger or gar. In the Hebrides, Gardhr is compli- 
cated by the Gaelic garradh (garden), a word 
borrowed from the English, and also by the 
native Gaelic word garadh, copse, den, which 
appears in the names of one or two glens (Glen- 
Garry). As in the case of dail, when Garry is 
prefixed the name is Gaelic. Names of farm* 
which appear to be Scandinavian are — Croi- 
garry, for Kraer-gardhr, that is, Kros' -garth; 
from Kro, a pen; here the place which at first 
was only a sheep-fold has become settled ; Asmi- 
garry, for Asmundar-gardhr, that is, Osmund's 
garth or farm; there is another Oshmigarry in 
Skye. The mutations to which the proper name 
is subject is shown by the Orcadian " Asmundar- 
vagr," which passes in 1503 to " Osmundwall," 
and at last appears as " Osnawall." Timi- 
garry may be Tuma-gardhr ; where Tumi is 
Thomas. Eusigarrv, Rushigarry, in Bernera, 


Harris, has been Hris-gardhr, Bush-girth; from 
Hris, a shrub, brushwood. Trumsgarry, in 
North Uist, may be for Thrums-gardhr, that is. 
Thrum's, or Slow man's garth. 

Gil, a narrow gh^n with a stream in ii, 
appears terminally very often. In vSkye we 
have rivers and streams named Varra-gill, Vikis- 
giil (Wick-gill), Oisgil (Oyce-gill), and possibly 
the place names Galtrigil (Hog's-gill) and Crisi- 
gill (Cross-gill) may belong to the same word. 
The word Udrigill, which appears so often in 
connection with points and capes, could easily 
be explained phonetically as Ytri-gil or Further- 
gill, if the physical features of the places always 
suited. In Arran, we have Scaftigill (Shaft- 
gill) and Catacol (Cat or Ship's gill); and in Islay, 
as a farm name, Giol. Tralagill in Assynt is 
doubtless Troll's gill. 

Gisl, a hostage, bailiff, and a person's name. 
The name is not found in the Orkneys, nor in 
Shetland, but it is not uncommon in Iceland 
(Gislakot, Gislabeer, etc.). Gisla, in Uig, Lewis, 
ivS certainly formed from this word, and may be 
a proper name. The terminal part may be a, 
water, or hie, shelter, or hlidh, slope. 

Gnipa, a peak. This word occurs in the 
Orkneys, in Gal-neap; Neep, in Shetland; Nipa, 
in Iceland. In this case, the Scandinavian form 
is better preserved in Lewis than in the northern 



islands, as it is written and pronounced Kneep, 
but its Gaelic form, Crip, is rather confounding. 
There is a Oreepe in Skye, not far from Dun- 

Gras, grass; pi. gros. We find Gresmark, 
Grashol, in Iceland; Grassfield, Girsigarth, in 
Shetland; Grassholm, Girsa, in the Orkneys; 
and Grasabhaig, a Gaelic form of Grasa-vik, 
Grasswick, in Uig, Lewis. The name of Gress, 
Stornoway, stands for gros — pastures; and a 
hint for the reason of the name may be found in 
Macculloch, who says, " A body of limestone 
occurs at Gres " (p. 194, Vol. I., Western Isle»). 

Grof, a pit. There are Grof and grafirgill, 
in Iceland; Graven, Graveland, Kolgrave, in 
Shetland; and Grawine, in the Orkneys. In 
the Hebrides there is Graffnose, in South Uigt; 
Grafirdale — erroneously spelt Cravodale — in 
Harris ; and Gravir, in Lochs of Lewis ; it stands 
for Grafir, pits, ravines. 

Heimr, an abode, a village. This word it 
rare in Lewis, but it appears to occur in Borsam, 
Harris; and the ancient form may have been 
Borrs-heimr or Bas-heimr. (Captain Thomas). 

Holl, a hill, hillock. This term, although 
not common, appears to occur in Arnol, Barvas; 
for Arnar-hoU, Orns-hill, where Orn, gen. 
Arnar, is a proper name, though the noun orn 
means an eagle, that is, the English erne. In 
Lionel, Barvas, we have Lin-holl, Flax-hill. 


Kjos, a deep, hollow place. There are 
several places named Kjos in Iceland; and 
Keotha, in Shetland, may be the same word; 
but it is surprising to find in Keose, Lochs, 
Lewis, the name so well preserved. It occurs 
again on the east side of Loch Seaforth. 

Klettr, a rock, chff. In the Orkneys, a 
precipitous, detached holm is called a Clett; 
while, in Lewis, clet is applied to any rou^h, 
broken-faced hill. It is one of the most common 
name* for a hill in Lewis. Inaclete is probably 
cognate with Ingyebuster, Orkneys; Ingasten, 
Shetland; and Einganes, Engamyrr, Iceland; 
for Engjaklettr, that is, clet of the meadow. 
Enaclete is also for Engja-klettr. Hacklete is 
certainly Har-klettr, high-clet. Breaclet is 
paralleled by Braebost, Skye; Breaquoy, Ork- 
neys; Breidharhlidh, etc., Iceland, and stands 
for Breidhar-klettr — that is, Broad-clet. Dira- 
clet, Harris, is cognate with Jura ; with Duirinish, 
Skye, Deerness, Orkneys. There are no Dyr 
commemorated in Shetland; for Dyra-klettr is 
for Deer's-clet; Dyr means an animal, t deer. 
Breasclet may be Breidhar-as-klettr — Broad- 

KoUr, a top summit. KolLr, in Iceland, is 
represeated by Coal, in Shetland, and, perhaps, 
t)y Colsettr, Orkneys. It is Coll, in Lewis; 
otherwise Koll, Kolle, for Kollr. 


Kross, a cross. There are ten places with 
this name, Kross, in Iceland; and three (Corse, 
Cross, Crose), in the Orkne3^s; and in Shetland 
it appears in various forms in combination. 
Besides Cross, in Barvas, there is Grossbost in 
Lochs, Lewis. Crossapoll, in various forms, 
has already been noticed in regard to Tiree, 
Coll, and Islay. 

Midi, a jutting crag. This as Mull, Moui, is 
in common use in the northern islands, and is 
not infrequent in Lewis; but it does not enter 
into the name of a farm except in Clashmeil, 
Harris, which may be Klas-muli, and cognate 
with Klasbardhi, Iceland. Joyce is wrong (p. 
383) in deriving Mull in the Mull of Galloway 
and Mull of Kintyre from Mael, Gaehc, a bare 
promontory; it is from Norse Muli, a high, bold 
headland, and not implying " bareness." Other 
mulls are the Mull of Deerness and the Mull of 
Papa Westray in the Orkneys; Blue Mull in 
Unst, Shetland; Mi^ilin (thrice), Faeroes; Muli 
(seven times repeated), Iceland. 

Nes, a ness or nose. Ness is a very com- 
prehensive topographical term, including not 
only the high chalk cliffs of Cape Grinez, but 
also the low shingle beach of Dungeness. It is 
usually written nis in Gaelic, and pronounced 
" nish." Sometimes "Ness" becomes not 
only the name of a " ness " proper, but of a 
large district. The Northmen invariably called 


the modern county of Caithness by the name of 
Nes, and the northern district of Lewis is known 
by the same name, Nes. There are ten farms 
called Nes in Iceland, and Ness occurs both in 
the Orkneys :uid in Shetland. In the Lewis 
Eental the entry is " Fivepenny Ness "; John- 
ston's map gives " Fivepenny " alone; and the 
Ordnance map translates the latter into Gaelic, 
" Cuig Peghinnean," Five Pennies. Aignish 
is called by the aU-observant Martin, " Eggi- 
ness " ; and he remarks : — " The shore of Eggi- 
ness abounds with little smooth stones, prettily 
variegated with all sorts of colours. They are 
of a round form, which is probably occasioned 
by the tossing of the sea, which in those parts 
is very violent " — (p. 10, West. Isles). In 
Captain Thomas's opinion also, Aignish was 
probably named from these egg-shaped pebbles, 
thus Aignis would stand for Eggia-ness, 
from Norse egg, an egg. But egg also means 
an edge, which equally well explains the 
name. Steinish is represented by Stein-nes in 
Iceland; Stennis, Orkneys; and Stennis in Shet- 
land. The decay of the great conglomerate 
has, around Stornoway, left great quantities of 
smooth, water-worn boulders and pebbles — 
hence Steinish for Stein-nes, from steinn, a 

Arinish, better written Arnish, has its 
counterpart in Skye (Arnish); as also Arnisort 


(where ort = fjordhi), occurring again in leeland 
as Arnarnes, Arnanes; from Orn, a proper name 
meaning eagle, the feminine of which is Orna. 
Captain Thomas thus refers to words in Kkr : — 
" Aaernish is repeated again in South Uist as 
Earnish, and again in Skye, where we have also 
Eaasay. Although there is no record of the 
roe-deer in Lewis, this name tells us that they 
were once there. Eaernish, otherwise Eairnish, 
is close to the Birken Isles, and *' roe ' are 
included in a contract for protecting the game 
in 1628 (p. 190, De. Eeb. Alb.). For Raar- 
ness means Eoe-deer's-ness; from Ea, a roe." 
Since writing the first part of these papers, Eev. 
Mr Mactaggart, of Glenelg, has drawn our atten- 
tion to the fact that Eaasay is sometimes heard 
pronounced Eaarsa, with an r. This spelling 
we knew of as existing in Dean Munro's work 
(1549), but thought it an oddity of his own. The 
word may mean " roe-island," though still the 
double genitive in rs is difficult to account for. 
We might look on ra as both masculine and 

We find the meaning of Breinish by com- 
paring it with the oft-repeated Brabuster, in the 
Orkneys, and Brebuster, in Shetland, which are 
contractions of Breidharbolstadhr, Broad-farm, 
of which there are ten in Iceland. Breinish, 
then, is for Breidhar-ness, Broad-ness. Gar- 
nish, Uig, appears again as Garnish, North Uist, 


and as Carness, in the Orkneys. It seems to be 
cognate with Kjara and Kjors-eyri, in Iceland; 
if so, Garnish stands for Kjarr-ness, Bushy-ness ; 
from Kjarr, copsewood, brushwood. Haroldsen 
has Kiorr, to mean palus; gen. kiarrar, terra 
gahiiosa, aquatica; this describes both the Car- 
nesses; but the word, in this sense, is not in 
Cleasby's Norse Dictionary. Callernish is an 
interesting name and place. It may have been 
Kjalar-ness ; from Kjolr, a keel, a keel-shaped 
ridge. But the fine Celtic megalithic cross-circle 
and avenue which stands upon the top of it sug- 
gests that the Northmen may have given to the 
point one of the names of Odin, viz., Kjallarr. 
Kjalar-nes is a place name in Iceland. Quidi- 
nish seems to be cognate with Quidamuir, and 
Quiderens, in Shetland, and is probably an 
abbreviation of Kviganda-nes, Quey-nes; from 
Kviganda, a young cow or bullock. Manish is 
repeated in Skye; and in the aspirated form of 
Vanish (in Gaelic, Mhanis) it occurs at Storno- 
way, Benbecula, and South Uist. In Iceland 
man, mana, is a common prefix (mana-vik, 
man-a), where mani is a proper name. " Aii 
intimate acquaintance with Stockinish enables 
me to give its etymology; it is Stokki-nes, Stakk- 
ness, from a chasm (stokkr), navigable at high 
water, which separates the island from the main. 
Stokkr means the narrow bed of river between 
two rocks; compare stok-land, an isolated rock. 


Molliiigiaish is in fact two words, Mol Lin<^inish;. 
wiiere Mol is the Oaelic for beach. Lingiiiish i* 
foi- Lyngar-nes, Ling-ness, Heather-ness ; from 
Lyng, heather. Hushinish, in Harris, occurs as 
Ku Ushinish (an iteration of the idea of cape) in 
Lewis, and again in South Uist; and cognate 
names are Husabost, in Skye; Housby, in the 
Orkneys; Housay, in Shethmd. Husa-nes, in 
Iceland, is identical, meaning ' House-ness ' ; 
from Hiis, a house." — (Captain Thomas). 
Orishernish in Skye is for Grisarnes, or pig's 

Oss, that is, oss, the mouth or outlet of a 
river, oyce. In Barvas the termination is, no 
doubt, oss, i.e., Barv-oss. The prefix may be 
an obscure form of Bara, a wave, billow; but 
we do not find any other name like it, and there- 
fore do not receive the usual help from analogy^ 
In Aros we have the combination of a (water) 
and oss; the noun aross means a river mouth or 
oyce in Norse. 

Papi, a pope, priest. The Scottic clerici, 
Celi-de, Servi-Dei, must have been bold and 
hardy seamen, for some of them sailed in the 
month of January, about the end of the eighth 
century (a.d. 795?), to Iceland, and stayed there 
till August, and when the island was colonised 
by the Northmen in a.d. 874, they found missals, 
bells, and crosiers at places on the south-east 
coast, which from that circumstance they called 


Papyli and Papay. One of these clerical sailors 
informed Dicuil that he had sailed (from Shet- 
land most probably) to the Faeroes in thirty-six 
hours, in a four-oared boat. This, as the 
distance is about 180 miles, would give the 
moderate rate of five miles per hour. At that 
time hermits had lived there for nearly one 
hundred years. There is neither Papyli nor 
Papa in the Faeroes, but they must l)e com- 
memoi'ated in Vestmann-hafn, though the name 
as it stands only indicates the former presence of 
the Gael or Westman. In Shetland, before the 
devastation by the Northmen, the Celi-de or 
Culdees were established in Papal, Unst; Papal 
Yell; and at Papal Burra; as well as on Papa 
Stour, Papa Little, and Papa, in the bay of 
Scalloway. In the Orkneys they were located 
at Papley, South Ronaldsha; Paplay, Holm; and 
Papdale, at Kirkwall; as also at Papa Stronsay 
and Papa Westray. In the Hebrides the Ceh- 
de are commemorated at Payble, North Uist; 
Papadill (papa-dalr), Rum; Paible, Harris; and 
Pyble (Byble, Bible !), Lewis. All these forms 
are variations of Papyli, which represents 
Papabyli, PapbyH, or Priest's-abode, one labial 
absorbing the other. Besides these, the Servi- 
Dei must have been established on Pabbay, 
Skye; Pabay, Barra; Pabbay, Loch Boisdale, 
South Uist; Pabay, Harris; and on great and 


little Paba}', Lewis; the original foini being 
Pap-ey, oi- Pi'iest's-island. 

Bif, a reef in the sea. Reef, as the name of 
a farm, occurs in Lewis, South Uist, and the 
Orkneys; and as Rif, in Iceland; in every case 
from an adjoining " reef." 

Setr (1) a seat, residence; (2) mountain- 
pastures, dairy-lands. This noun, so common 
in the names of farms in the noi'thern and 
western islands, is not to be found at all among 
the seven thousand in the Icelandic Ny Jardhar- 
bok. In the Orkneys and Shetland the 
" setters," which originally were only summer 
" seats," have become fixed residences and 
cultivated lands. In Lewis, in mid-summer, the 
home farms are almost deserted, the men being 
at the herring fishing, and the women and cattle 
on the moors. There are thirteen " Shadirs " 
named in the Lewis Rental; when written in 
Gaelic the word is Seadair, pronounced shader. 
There are places of this name in Lewis; Bernera, 
Harris; and in Skye. In the Orkneys we have 
seater; in Shetland, setter; in Landnamabok, 
saetr. Some of the differentiated " setters " of 
Lewis can be readily resolved. Grimshader is 
identical with Grymsetter in the Orkneys, 
Greemsetter in Shetland; and cognate to Grim- 
stadhir, in Iceland. Grimr is a very common 
Scandinavian proper name, and the leaPRed 
editor of the " Icelandic Dictionary" would fail 


persuade us that it by no means implies an un- 
amiable person. Grimshader, for Grim-setr, 
means Grimm' s-setter, seat or pasture. Ker- 
shader is met with as Cursetter, in the Orkneys ; 
for Kjor-setr, that is, Copse or Brushwood 
setter. Besides Quishader, in Lewis, there is 
Quinish, Bernera, Harris; Vallaque, North Uist; 
and the far-famed Cuidhrang, in Skye — a Gaehc 
spelHng of quoyrand, Kvi-rand, round-quoy. 
" In the Orkneys," says Captain Thomas, 
" quoy is a subsidiary enclosure to the principal 
farm, and is the only exception I know of to the 
rule v/hich governs Scandinavian names, by 
being used as a substantive prefix. Sometimes 
a quoy is only a few square yards of land, 
enclosed by a rough stone wall, to rear and pro- 
tect young cabbage plants; this, in Shetland, 
would be called a cro. In Shetland we have 
' Queys, Quiness,' etc., but the name is not 
common; in Iceland, Kvi-bol, Kviar-ness. 
Quishader, for Kvi-setr, fold or pen setter; 
from Norse Kvi, a fold, pen. Earshader has 
cognate representatives in Air, Irland, in the 
Orkneys; Erebie, Sandsair, Ireland, in Shet- 
land; and Eyri; Eyarhus, in Iceland. Ear- 
shader, for Eyrar-setr, Beach-setter, from Eyrr, 
a gravelly bank, beach. Linshader is the 
embarking place for crossing to C^allernish, and 
may very well be Hlein-setr; from Hlein, a rock 
projecting like a pier into the sea ; but it is more 


probably Lon-setr, that is, Creek-setr; from 
Lon, an inlet sea loch." Shiilisheder, which 
appears in the Long Island and in Skye, may be 
from Sula, a |)illar, a root word which appears ia 
Sulisker islet (Pillar-skerry), and in the Assynt 
mountain name of Sulvein, the pillared hill. 

Other shaders are Limshader, Sheshader, 
Gurshader, Carishader, Geshader, and Ung- 
shader, but their interpretation is not easy, for 
the prefixes may be variously resolved. 

Uigshader, in Skye, means Wick or Bay 
Seat; and Ellishader, the Ellister of Shetland, 
may stand for Hellis-setr, or Cave-seat. Ard- 
elester and Ellister, in Islay, have similarly 
been explained. 


Gaelic and Norse differ widely in their methods 
of combining compound words so as to make 
up a place-name. Gaelic places the possessive 
genitive second, and the generic term (town, 
village, hill, field, etc.), first; Norse places the 
genitive first. It is similar with the qualifying 
adjectives — Norse, as English, places the adjec- 
tive before the noun; Gaelic places it after. 
This helps us immensely in deciding upon the 
Norse or Gaelic character of a place-name. For 
instance, if val or mheall comes after the parti- 


i^ular possessing' noun or the qualifying adjec- 
tive, we deal here with the Norse fell and not 
with the Gaelic meall. Thus, Griomabhal (grim- 
fell) would, in Gaehc, be Meall-gruamacli. 
Gaelic has borrowed one or two of the Norse 
generic terms, notably dail (dale) and sgeir 
(skerry or insulated rock). When dail or sgeir 
comes first, the combination is of Gaelic origin ; 
when these words come last we deal with a Norse 
name. Thus we have Dal-more and Skerry- 
more, the big dale and skerry, as against Swoi- 
dale (Sward-dale) and Ilasker (High Skerry) of 
the Norse. We must be specially careful, in 
dealing with dal and sker, to remember their 
position in the compound. 

The above remarks and cautions have been 
necessitated by the fact that in pursuing our 
alphabetical list of Norse generic names in our 
insular topography, we have come to the word 
Sker, which so many people who deal in etymo- 
logy fancy to be a native Gaelic word. It is the 
Norse sker, a skerry or rock. It is a common 
word in place-names all round Britain, applied 
to rocks and skerries ; but as entering into farm 
and town-land names it is rare. We have Vati- 
sker in Lewis, which Captain Thomas referred 
to the adjoining Vadha-sker, or dangerous 
skerry. The famous Talisker is probably the 
hall of the rock (Norse hollr, a hall). 


Stadhr, a "stead" or abode. This word 
appears in the Enghsh steading and homestead. 
In Iceland stadhr or stadhir forms the termina- 
tion of 61 local names in the old Landnamabok.' 
In local topography in the northern Isles it 
means the place on which the dwelling stands. 
In Shetland, by 1576, stadhr had usually been 
shortened to sta. This is frequently now 
changed to ster. 

In Earl Sinclair's rental of his share of the 
Orkneys (1502), which in part seems to have 
been copied from an older document, stadhr is 
represented by " stath," " stayth," " staith." 
By 1595 " staith," " stayth," had been reduced 
to "sta," but a real corruption was introduced 
by " stane," and this has now generally become 
" ston," " ton," " tonn." We can trace the 
whole change in Grims-stadhr, which in 1503 
appears as Grymestath; in 1595, Grymston and 
Grymestan, and which is now written Gremiston. 

In Lewis stadhr is not an uncommon generic 
term. Skegirsta — the Gaelic form of which is 
Sgiogarstagh — is the same name as Skeggja- 
stadhr in Iceland and Skeggestad in Norway, 
ai4d indicates that Skeggi was located there. 
Mangarsta, occurring as Mog-stat, Mugstot, 
Monkstadt in Skye, and as Mangaster in two 
places in Shetland, was Munku-stadr, and tells 
us that it was formerly the abode of monks. 
Meahsta is Melastadr; from Melr, i.e., sand-hills 


overgrown with bent grass, in Scottish '' hnks." 
We have Melbost twice in Lewis ; Melsettr in the 
Orkneys, and Melby in Shetland. In Iceland 
there is Melar, and the same name as m Lewis, 
Mel-stadr. All these places are sandy, and 
in sunmaer luxuriantly green. The monks of 
Mangaster may have joined in spiritual joys with 
the Oailleacha Dubha, i.e., nuns, the site of 
whose house is still to be seen at Mealista. 

There are two " Tolsta '* in Lewis, which 
may have been Tolu-stadhr, that is, Toh's-stead, 
of whom seventeen are named under a great 
variety of spelling as pilgrims in the Reichenau 
Obituary; but it is strange that neither in Ice- 
land, Shetland, nor Orkney, is any name like 
Tolsta found. This would suggest that the name 
may really begin with h, and be Hol-stadhr, 
hollow or low stead. Crowlista or Crolesta may 
be for Kro-hljidh-stadhr, or Pen-lea-stead, but 
we cannot be certain. Borrowston is possibly 
Borgar-stadhr, Burg-stead; the ston originated 
in the same manner as the Orkney names from 
Btadhr considered above. There is no tun, 
town, in Lewis, and it is rare in Iceland, being 
appiied to insignificant places, and equally rare 
in Orkney and Shetland, where false analogy and 
En^^ish influences modified the sta. 

In Harris we have Scarista, and there is 
another Scarista in Uig, Lewis, not named in the 
Rental; these are synonymous with Skara-stadr, 


in Iceland. Skari (skorey, in Shetland) is a 
young gull still in its grey plumage; but it is also 
a nickname, so that Skara-stadr if* not the 
'•' stead of a skorey," but the " stead of Skaii."" 
This word skari is borrowed into Gaelic as 
sgaireag with a like meaning (see Mr M'llury'e 
interesting remarks on this bird in the 3rd 
volume of the Highland Monthly, page 353). 

Erista in Uig adjoins some quicksand which 
has been fatal to horse and driver, and Captain 
Thomas suggests the root as yrja in sand-yrja, 
quick-sand, adducing the Icelandic place-name, 
Irjar. But it is likely the same root word as in 

Strond, a strand, coast. It is represented by 
Strond in Harris, Strand in Shetland, and Stiond 
in Iceland. Strandabhat appears twice as a 
lake-name in Lewis; it means strand water. 

Troll, a giant, troll; trijlla, to enchant; 
tryllskr, bewitched. Ballantrushal, properly 
Baile anTruiseil, a township in the west of Lewis, 
which takes its name from Cloch an Truiseil, the 
Trusel-stone. This is a gigantic monolith or 
standing stone, which, as Captain Thomas re- 
marks, well deserves the title of Tryllskar-steinn, 
that is, the stone of enchantment, and which 
has become Tryskall, Tryshall, by metathesis. 
Doubtless, in Trysil Fiall in Norway the same 
form is seen. The legend connected with the 
stone is best related in another place. The 


sagacity of the topographer is sometimes severely 
taxed — " L Vnsal Sago " is not to be directly 
recognised " Trusal Stone," nor does "B 
Trade ' ' immediately suggest ' ' Baile an 
Truiseil." That is how maps are made. 

Toft, a knoll, or toft or tuft. This is bor- 
rowed into Gaelic as a common noun in the form 
of tobhta, or tota, and it generally means turf. 
It appears in Lewis as Totta, Totaichean Aulaidh 
(Oiave's Tofts), and in Skye in Totscore. 

Tung a, a tongue, tongue of land. This is 
a very frequent place-name in Northern and 
Western Scotland. It is frequent in Iceland, 
and occurs as Toung in both Orkney and Shet- 
land; it is Tong in Lewis, and Teangue in Skye. 

Vdgr, a creek or bay. It is a troublesome 

though most important word in topography, for 

both its V and // may disappear in Gaelic. In 

Iceland the word is clear and easy ; it is common 

as a place suffix. In Shetland vagr retains iti 

right sound as " voe," except only in Scalowa, 

Scaloway, the final way being a concession to 

folk-etymology working on the English " way.'' 

Compare Bible Head (Paible), and Europa Point 

of Lewis. Scaloway stands for skala vagr, 

shieling or hall bay, skali being a hut or hall. 

The Orkneys have few bays or voes; but here 

there is a great confusion between vagr and 

vollr, a field, for both are made to end in "wall" 



now; nor is the confusion lessened by the fact 
that voe and field are almost always adjacent. 
Many farms in Orkney end in -wall. One of 
them, Bigswall, is not near the sea, so that vagr 
is out of the case. Nor can we doubt as to 
Green-wall being Green-vollr, that is, Green- 
field; or Ting-wall, meaning Thing-vollr. But 
there is a parish Walls in Shetland, and another 
Walls in the Orkneys, both of which are histori- 
cally Vagarland ; Osmandwall is Asmundar-vagr ; 
Widewall is Vidhivagr; and Kirkwall, originally 
Kirkjuvagr, continues as late as 1525 as Kirke- 
vaag. It is evident, therefore, that in the 
Orkneys -wall may represent either vagr or vollr. 
On this point Captain Thomas says : — "Yet it is 
with extreme reluctance that I yield to this con- 
clusion; there is no difficulty with wall from 
vollr, but how, I ask, could vagr come to be 
represented by wall? From whence came the 
II? Was it that Scottish immigrants finding the 
sound of vd represented it in writing by ' wall,' 
the II at first being silent? But the opinion I am 
inclined to adopt is that both forms were current ; 
as noted above, where ' wall ' represents vagr, 
a vollr is also present. Besides Kirkju-vagr, 
there was always in fact Kirkju-vollr (Kirkfield), 
Kirkwall; and so of the rest. And the parish 
names Walls appear to me to be used m contra- 
distinction to the peculiarly mountainous dis- 
tricts of Sandness in Shetland, and Hoy in the 


Orkneys; if so they would have been called 
veUir, Englished by Walls (Vales). It is true, 
however, that both ' walls ' are largely inter- 
sected by ' voes.' The solution of the question 
depends upon whether most weight is given to 
the induction from observation, or to the 
historical documents." 

In the Hebrides this unfortunate word vagr 
is plagued by complications of another kind. 
The Rentals, indeed, record the names of farms 
with greater purity than in the northern islands, 
but they have been written by Northern Saxons; 
V is turned to w, and vagr becomes " way " ; no 
doubt, when first written, " way "' rhymed with 
"far," but now, in common English speech — 
from the influence of the written form — it 
rhymes with " day." But in the native (Gaelic) 
speech, no word can have an initial v in the 
nominative case ; also if two nouns are combined 
to form a word, the suffix, if capable, suffers 
aspiration. These rules are sometimes strictly 
followed and sometimes not. We will take the 
examples of Carlo way, which undoubiediy was 
Karla-vagr, that is, Carl's bay. In Karla-vagr, 
the final r in Norse merely emphasised the pre- 
ceeding consonant. When the Gael took posses- 
sion of the word Karla-vag, they would do, as the 
northern islanders have done, viz., drop the g, 
and next they would consider va' to be a noun 
in the genitive case, and would therefore soften 


the d to ai, thus sounding " vai," which in Gauhc 
orthography would be bhaidh, bhai.i^h, and of 
which the nominative w'ould be bagh, badh; this, 
again, translated into English, w-ould be " bay." 
In this roundabout manner the vagar of Harris 
have become the " bays." Karla-vagr, reduced 
to Karla-vai, would be written Carlabhaidh, and 
Teutonic influence, changing the v or hh to w, 
brings us to Carlo-way. 

Stornoway : this name is repeated (Loch 
Stornua) in Kintyre. In Iceland there w^ere 
formerly Stjornu-stadhr and Stjornusteinar, 
but these names are now obsolete. Storno- 
way — which is spelt in thirteen different 
w-ays — has been referred by Captain Thomas 
to Stjornu-vagr, Star's-voe; where Stjarna, 
Star, is a proper name. " The only person 
I find," adds Captain Thomas, '"recorded 
bearing that name is Oddi, who was so 
learned in astronomy that he was called ' Stjornu- 
Oddi.' He had a remarkable dream, ' Stjornu- 
Odda Draumr,' but it appears to have had 
nothing to do with the stars (Nordiske Old- 
skrifter, XXVII.). It may be noted that no 
place on land in Kintyre bears the name of Stor- 
noway, which disposes of the foolish Gaelic 
etymology of Sron-a-bhaigh, Bay-nose." This 
derivation of the Captain's is unsatisfactory. 
The root word clearly is Stjorn, steering, the 6 
of which is stable, and does not change to a. 
Stornoway, we take it, means " Steerage-bay." 


Stimeravay stands for stemdi-vagr, the 
stopped-up voe, which describes the place. For 
Carloway, Blaeu's Atlas has Carleywagh; there 
is no cognate in the Orkney or Shetland Isles, 
unless Charleston in Aithsting. In Iceland there 
i» Karlafjordhr, that is, Carl's firth, where Karli 
either means carle or a man's proper name. 
Flodeway either stands for Fljota-vagr (stream 
or flood voe) or Flota-vagr (Fleet's voe). 

Vatn, water, lake. This is by derivation the 
same word as English water and the Greek 
hudor, hydrant, etc. It is very common m 
Hebridean place-names, where it appears al 
bhat or vat. The lake named Langavat or Long- 
water appears several times; and there are 
Breidh-bhat (Broad lake), Skara-bhat (Skari's 
water, see Scarista), Lacsabhat (Lax or salmon), 
and numerous others. An interestmg perver- 
sion of vatn appears in Loch Sandwood, in 
Eddrachillis parish. 

Vik, a creek, bay, wick. This appears ter- 
minally in Gaelic as bhaig, aig, or ag. A 
common name is Sandwick, that is. Sand-bay. 
It is in Iceland, Shetland, and Orkney, in the 
Long Island, and as far south as Arran, where 
we have Sannox, a plural form to denote that 
there are three sand-bays there. Marweg or 
Marvig appears as Marwick, in the Orkney and 
Shetland Isles. It comes from mar, a sea-gull. 
The same name may also appear in Maraig or 


Marag, which Captain Thomas erroneously 
refers to vagr, voe. The prefix he explains as 
myrar or mire (bog) voe. Meavig or Meavag, in 
Lewis and in Harris twice, is for Mjo-vik, narrow 
\'0e. Kerriwick, otherwise Kirvis. has been 
referred to Kirkju-vik, that is, Kirk-wick. 
Colivick goes along with Crowlista, explained 
under stadhr. There is a Cruely in Shetland. 
The simple form of Vick appears often ; we have 
the town of Wick in Caithness. There are one 
or two parishes called in Gaelic Uig (Lewis and 
Skye), that is, Wick. Captain Thomas strangely 
refers these to the Norse ogr, an inlet or creek, 
which is a rare name, not found in Orkney or 
Shetland, and twice in Iceland. 

We have now passed in review the principal 
Norse words that enter into the place-names of 
the Hebrides. The universality of these names 
in the Long Island is most remarkable. In con- 
sidering the Gaelic names, we shall find that the 
Norse names beat them in Lewis by four to one , 
and, further, these Gaelic names are importa- 
tions since the re-occupation of the Islands by 
the Gael on the fall of the Norse power. 







Inverness as a town name goes back to the 12th 
century, possibly further — to Macbeth's time 
(1057). The name is partly, at least, of Gaehc 
origin, which proves that the town could not be 
so called in the time of the Picts, who would have 
named it Aber-ness, not Inver-ness. The word 
hibhir (Inver) means in Gaelic a confluence, and 
is by derivation the same word exactly as the 
English word " infer." The town derives it» 
name, of course, from being at the confluence 
of the Eiver Ness. The name Ness appear* in 
Adamnan's. life of St Columba as Nesa; the 
Norse called it Nis; and the modern Gaelic pro- 
nunciation and spelling are the same, viz., Nis. 
The word must be referred to an original form 


rtesta according to Celtic philological laws, and 
this Dr Whitley Stokes has equated with the 
Sanscrit word nadi, signifying river. We may 
compare also the old Thracian river name Nestos, 
and possibly the mother of the great mythic King 
of Ulster, Nessa, he being Conchobar Mac Nessa, 
bears the same name, for she may have been a 
river goddess. Rivers were worshipped as 
deities, as we know from Gildas, and as such 
names as Dee and Don (Deva, Divona), meaning 
" goddess," prove. 

The name Clachnacuddin stands for Clach 
nan Cudainn, or stone of the tubs, a phrase 
which is explained as referring to the habit of 
the women carrying water from the river, and of 
resting their tubs on the stone that now forms 
the Palladium of Inverness. Markinch stands 
for Marc-innis, the horse isle or " inch." The 
name is interesting in two ways; firstly, animal 
names may come before the word innis, although 
Gaelic otherwise insists on all genitive or pos- 
sessive forms coming last; secondly, the word 
trims, so common all over the country in place 
names, is now obsolete in the sense of " island," 
its only meaning in present Gaelic being a shelter 
for cattle, such as a clump of wood and the like. 

The Haugh and Holme. — The name Haugh is 
English, or rather Scotch; the word originally 
possessed an /, now lost; Barbour in his Bruce 
writes halche, and the Anglo-Saxon form is 


healh. The Haugh is mentioned in a charter, 
1361, as Hale; and the Gaelic people still keep 
up the old / in their version of the name, that is, 
an Talchan. In borrowing English or Scotch 
words beginning with h, the Gaels always intro- 
duced a t; so we find the Gaelic of Holme, so 
named from the English holm, an island in a 
river, to be Tuilm. 

Ballifeary. — This appears in 1244 as Balna- 
fare, and it is explained as the Gaelic Baile-na- 
faire, the town of the watching. Similarly, 
Clachnaharry means the stone of the watching. 
The town had to place sentinels at these points 
to give notice of any hostile visit which the rest- 
less clans around might think proper to make to 
the town. 

Drummond. — This name is common all over 
the country. It is the locative case of the word 
druim, a ridge, which had a stem ending in men 

Bught. — This is the Scotch word bought, 
houcht, signifying a bending, a bay, a pen. The 
root is the Teutonic hugan, to bow, which is 
possibly alhed to the Gaelic word hog, whence is 
borrowed the English word hog. 

Kinmylies. — This appears in 1232 as Kin- 
myly, and it has been well explained as standing 
for ceann mile, mile-end. 

Leachkiyi. — This is the Gaelic word leacuinn, 
a. face, hillside; it is really, like Drummond, an 


oblique case from a nominative leac, signifying 
cheek or face. 

Torrean and Kilvean. — Tiiese are the tor or 
hill and the kil or church of St Bean, a saint of 
the Celtic Church, usually reckoned first bishop 
of Mortlach (11th century). The name Bean, 
in Gaelic Beathan, is a derivation of the word 
beatha, life; and it has the same force as the 
more famous name Macbeth. The saint usually 
supposed to be meant here is Baithene, St 
Columba's successor, but in modern Gaelic his 
name would be Baothan. Dunain means the hill 
of birds (dun-ian). 

Tomnahurich. — Much fanciful nonsense has 
been written about the meaning of this name. 
The favourite derivation is that which refers it to 
the idea of " boat hill," for the mound looks like 
an upturned boat. The Gaelic may be written 
Tom na h-iubhraich; and the dictionaries give 
the word iuhhrach as meaning a boat. This, 
however, is nonsense; there was only one boat 
called the lubhrach, and that was the mythic 
boat of Fergus MacRo, in which he took over to 
Ireland the sons of Uisnech and Deirdre. The 
word iuhhrach simply means a yew wood, and 
the root word has given some famous place 
names. The word appears in Gaulish as Eburos, 
and hence we have the British Eburacum, which 
is now York (" Yew-town "). 


Dochfour. — The first part of the name — Doch 
— is easy ; it is the contracted form of dahhach, 
a tub, and then a measure of land equal to four 
plough-gates. The latter part — four — is an 
extremely difficult term to unravel. It appears 
in many place names, but only in Pictland. We 
meet with Balfour, Pitfour, Delfour, Tillipowrie, 
Letterfour, &c. It is a maxim in deriving these 
names that the chief term lies where the accent 
is on the word, and four always carries the 
accent. This settles that it is a noun and not an 
adjective; for some people will have it, despite 
Gaelic phonetics, that /our is simply the adjective 
fuar, cold. The preserved / shows that the 
word began with p, and this again proves the 
Pictish character of the word ; for p is not native 
to Gaelic. I have elsewhere suggested that this 
word is allied to the Breton peur, Welsh pawr, a 
pasture — a derivation which Dr Whitley Stokes, 
one of our best Celtic philologists, has accepted 
most cordially. 

Ahriachan. — This appears in 1239 as Abir- 
hacyn, and in 1334 as Aberbreachy. The name 
undoubtedly stands for Aber-briachan, the con- 
fluence of the Briachan. For the loss of a 
syllable between aher and briachan, consider 
Arbroath and Aberbrothock. One of the hers 
simply has gone. The stream running into Loch 
Ness at Abriachan must have originally been 
called Briachan; it has now two or three names. 


mostly descriptive of the water — Allt-dubh, Allt- 
dearg, Allt-liath. What Briachan may mean it 
is hard to say; it is evidently Pictish. The 
derivations offered for this place-name (Abri- 
achan) have been very numerous ; a full account 
of them is given in the third volume of the 
Transactions of the Inverness Field Club just 
issued (pp. 167-171). 

In passing beyond the bounds of Inverness 
and Inverness Parish, a word may be said about 
Bona, the alternate name to Inverness — the 
parish of Inverness and Bona. In 1233 this is 
spelt Baneth, and two hundred years later as 
Bonacht, Bonoch. It has been explained vari- 
ously as Ban-achadh, white field, and Ban-ath, 
white ford. It seems to me that it is a reminis- 
cence of Ptolemy's Banatia, which some authori- 
ties place here, and possibly is due to some 
pedant of the twelfth century. 

Petty. — This parish name, in Gaelic Peitidh, 
is a plural form of the well-known prefix pet, so 
common in Pictland, and meaning a farm or 
township . 

Doves. — This name must be compared to the 
parish name of Durris in Kincardine, which is 
nearer the Gaelic pronunciation in form than 
Dores is. The name seems allied to the word 
dorus, a door. 

Stratherrick . — This is named after the river 
Farigag, and the name Farigag probably stands 


for Far-gag, " above the cleft or rift," made up 
of the preposition far or for (above) and gag, a 
cleft or pass. 

Urquhart. — This is one of the oldest names 
in the district. Adamnan (700 a.d.) mentions 
the glen as Air-chartdan. The first part, air, is 
the preposition air, on. The significant part is 
cartdan, or the later cardan, cardainn. This 
last part appears in two other combinations, viz., 
Kin-cardine and Plus-carden. The name is 
undoubtedly Pictish, and therefore we have to 
look at Welsh, as the nearest of kin to Pictish, 
for the explanation of the name. The only likely 
one is cerddin, which means the rowan tree. 
Kincardine might thus mean the " end of the 
•rowan wood"; Urquhart, "by the rowan 

Daviot, Moy, &c. — The name Daviot is not 
unique; it appears in Aberdeenshire as a parish 
name. The Gaelic is Deimhidh, pronounced 
Devidh, Devy nearly, and answering the old 
charter form Deveth very well. It seems the 
exact modern equivalent of the old Welsh tribe- 
name Demetae, now Dyved; and here again we 
meet with an old Pictish name. The root dem 
signifies "fixed," "sure." Moy is in Gaelic 
A' Mhagh, "the plain"; while Dalarossie is 
simply Dail-Fhearghuis or " Fergus's dale." 

Ardersier, Croy. — The former appears in 
1227 as Ardrosser, and in 1570 as Ardorsier; 


the Gaelic, influenced by folk-etymology, is Arcl- 
na-saor, the cape of the carpenters. It has been 
explained as Ard-ros-iar, the west cape. The 
name Croy signifies hard, being from cruaidh, 

Beauly .—This is a mediaeval French name — 
heau lieu, pretty place — introduced by the ValHs 
Caulium monks in 1232, who founded the Priory. 
The Gaelic name is A' Mhanachainn, which may 
be rendered the " Monkery," from manach, 
monk. Kihnorack is the parish name; this 
means the church of Morag, that is, St Moroc. 

Foyers. — This is the name of the land below 
the falls by the side of Loch IS ess. It stands for 
Fo-thir, that is, "low ground." O'Eeilly says 
fothir signifies "good land," but the former 
meaning is doubtless the correct one here. Fort- 
Augustus in Gaelic is Kill-Chuimein, the Kil of 
St Cummin, a saint name which occurs more 
than once, the first being Cummian the Fair, 
who, about 650, wrote a biography of St 

The county names of Nairn and Ross may 
claim our attention for a little. The name Nairn 
primarily applies to the river Nairn, and this has 
invariably been connected with the woixl fearna, 
alder wood, despite the phonetic and other difii- 
culties. The names of large rivers and leading 
features of the country are the oldest of any; 
they go back to the times of other races and 


different languages. Names like Dee, Don, Tay, 
Ness, Nairn, &c., cannot be etymologised from 
modern Gaelic. We must rise to a hieiier level, 
and consider the Celtic tongues as a whole, even 
falling back upon European root forms to help 
us. And beyond all this, the name may have 
been borrowed from a primitive race that used 
a language that was neither Celtic nor Indo- 
European. The name Nairn is Celtic, for the 
termination jm is peculiarly so. The root is 
therefore na, possibly nav, which possesses the 
idea of sailing, swimming, floating, and is found 
in the English word naval and the like. The 
name is Pictish doubtless. So also is the name 
Eoss; this is alhed to the Welsh word Rhos, a 
plain, mead, which is very common in Welsh 
place names. 

Dingwall. — This is the Norse Tiling voUr, 
field of the thing or meeting — the meeting or 
paramount place. The name appears in the 
Isle of Man and in all places under Norse sway 
now or heretofore. 

Strathpeffer. — The river or streamlet named 
Peffer or Peffery appears three or four times in 
Pictland; we have two in Haddingtonshu^e, an 
Inverpephry in Perthshire, and this one at 
Strathpeffer. The Gaelic of Strathpeffer is 
Inbhir-feo'arain ; the Feofharan is the stream 
name, where the first / represents the Pictish 


original /), and the second /, now aspirated, 
either an / or more probably a b. There is a 
modern Welsh word pefr which might suit 
phonetically, meaning "fair," "beautiful." 

Ben Wifvis. — This, in Gaelic, is Beinn Uais, 
the oblique or locative form of Beann Uas; the 
word uas is now obsolete, but it was common 
once as a personal appellative — Colle Uas being, 
for instance, King of Ireland in the 4th century, 
and ancestor of the Macdonalds. The word 
means " noble," " majestic," and is but a 
shortened form of the Gaelic word uasal, proud. 





Thb County of Inverness can boast neither of 
symmetry nor of compactness. It sprawls west- 
wards across the northern neck of Scotland 
through Skye, dwing under sea to re-appear as 
the far-west s^a-bank of the Outer Hebrides. 
One thing it can boast of, however, among 
Scottish counties : it ig the largest of them. Its 
area of 4232 square miles — a square land-piece 
of 65 miles per side — is unsurpassed by any 
other county in Scotland. And once the Sheriff- 
dom of Inverness extended still further. In the 
twelfth century it comprehended all the country 
north of the Grampians, but the thirteenth 
century saw the rise of the shires of Elgin, Nairn, 
and Cromarty. For four hundred years there- 
after, however, the Sheriffdom of Inverness 
included Ross, Sutherland, Caithness, and part 


of Argyle. The present Sheriffdoms of Argyle, 
Sutherland, and Caithness were constituted m 
1631-3 and Ross in 1661, the latter three being 
pure dismemberments, so to speak, of Inverness 
Sheriffdom. The County of Inverness was thus 
finally formed in 1661 curiously by a process of 
subtraction, but it has kept its then acquired 
bounds ever since, with certain small adjust- 
ments. The irregularity of its northern borders 
from Harris to Beauly is due to the Mackenzie 
influence in 1661; that family wanted the clan 
estates to be all in Ross-shire. A scientific 
frontier was, therefore, out of the question. 

The history of Inverness county is nearly as 
sporadic in its character as the county itself. 
There is a separate story for the Isles, a second 
one for the west coast mainland (Garmoran), and 
a third story to tell of the province of Moray 
portion of the county. It is really a great pity 
that the old province of Moray itself was not 
made a county — a pity historically, for it was an 
ecclesiastical and almost a political unit. It 
included all Inverness east of the Drumalban 
watershed east of Lochaber, and comprehended 
also the shires of Nairn, Elgin, and even a part 
of Banff. Macbeth' s family province of Moray 
further included Easter Ross, disputed with the 
Norsemen, and its sway at times (11th century) 
extended over Banff and Buchan, as we can 
see from the Book of Deer. In the 12th 


century the old Earls of Moray were sup- 
pressed, and native thanes, with incoming 
Normans, began to take their place (early 13th 
century); the coast began to be planted with 
burghs. The great family of Gumming rose to 
power in Buchan, and early in the 13th century 
they acquired Lochaber and Badenoch. The 
Earldom of Moray was again restored by Bruce 
and given to Eandolph, his nephew, inclusive of 
Lochaber. The Church also occupied vast and 
valuable property in Moray, but the after history 
of the Moray portion of Inverness concerns the 
rise of the Gordons and their struggles with the 
Earls of Moray and the native clans, and 
scarcely bears on the place-names, which by 
this time were mostly fixed. The West 
Coast portion of Inverness-shire, north of Mor- 
vern, and extending to Glenelg — that is, 
Moydart, Morar, and Knoydart — was called 
" Garbh-mhorbhairne," in 1343 Garwmorarne, 
the " Garmoran " of the historians. It and 
Lochaber formed part of North Argyle, which 
once extended to Lochbroom. Garmoran be- 
longed to the descendants of Somerled of the 
Isles, a side branch (probably junior) to the Clan 
Donald. The heiress of Garmoran married John 
of Isla in the 14th century, and the property 
came to the Clanranald branch of the Mac- 
donalds. The Outer Hebrides belonged to the 
Norse, and therefore to the King of Man and the 


Isles; but after 1263, the date of the overthrow 
of the Norsemen, Skye and the Long Island fell 
as his share of the booty to the Earl of Ross. 
Forfeiting them in the wars of David II. and 
Edward BaUiol, he recovered only Skye, the 
outer isles going to his rival, the Lord of the Isles. 
The Island Lord next century succeeded also 
to the Earldom of Ross, sometime after Harlaw. 
This Prince therefore held (say) about 1450, 
through himself or his kin of Clanranald, all the 
Outer Hebrides, Skye and its adjacent isles, Gar- 
moran and Lochaber (inclusive of Glengarry). 
Glenelg belonged to his vassal, Macleodof Harris. 
On the break-up of the Lordship of the Isles 
(1475-1493), the local chiefs came to the front 
— Macleods of Harris and Glenelg, also of Dun- 
regan, Macneills of Barra, Camerons of Loch- 
aber, and the numerous but powerful branches 
of Macdonald — Clanranald (Garmoran and Uist, 
with the Glengarry branch further east, soon to 
lucceed in Knovdart another set of Macdonalds)^ 
the Clan Hugh of Sleat, whence the present Lord 
Macdonald, and the disinherited, because ille- 
gitimate, Macdonalds of Keppoch, in Brae 
Lochaber, whose lands were given to Mackintosh. 
The after history of these clans does not concern 
our subject; the place names with which we have 
to deal were given by the earlier clans, tribes,^ 
and races which had successively possessed the 
land prior to the 15th century. 


The earliest Celtic nation that established 
itself in Scotland was the Pictish. They found 
before them another race or two, one of which 
was fair and square-headed, and the other dark 
and long-headed. The Celts arrived in their 
iron age, possibly in 600 B.C. The language 
spoken by the previous inhabitants is unknown, 
the Picts spoke a dialect of Celtic near akin to 
the Welsh. Some Inverness County names bear 
out this fact. The test letter between the Brit- 
tonic and Gadelic or Gaelic branches of old Celtic 
is the letter p; old Gaelic had no letter p, and 
modern Gaelic developed native p within the last 
five or six hundred years; the many borrowed 
p's in Gaehc do not here count. Gaelic cuid is 
in Welsh peth (for older pett), a thing; this is 
the Pictish pet or pit, a possession or farm — ia 
«hort, the Gaelic haile in meaning. Here Pictish 
and Welsh show p as against Gaelic c, which, so 
far, proves Welsh and Pictish closer allied than 
Gaelic and Pictish. The pits or pets in Inver- 
ness-shire are not now so numerous as once they 
were. We have still Pityoulish (Abernethy), 
Pitchirn (Eowan-ton) and Pittowrie (Alvie), and 
Pitmean (Middleton, Kmgussie), and Pettyvaich 
(Byre-ton) in ICiltarlity. Balmaglaster of Glen- 
garry was formerly Pit-maglaster or Pitten- 
glassie. Several are obsolete — Pitkerrald (St 
Cyril's Croft) in Glen-Urquhart, and Pitchal- 
man and Pitalmit in Glenelg. Then there is 


Petty, the Parish name, which simply means the 
" land of farms " or " pets." Pet or Pit hai 
given way to its equivalent in meaning, haile, 
for two good reasons — the word first, like aher, 
was getting obscure, as not belonging to the 
ordinary vocabulary; and second, it got mixed 
up with another word of nearly like sound but 
obscene meaning. This especially has driven it 

Another test word is aher, a confluence; the 
Gaelic is inhhir or wrer (root her: in-fer); 
the Gaelic ahar, now obsolete, having meant 
a " marsh " (root of tohar). The Pictish 
aher had two dialect forms — aher and oher; the 
latter alone has survived in modern names as 
spoken in Gaelic — Obair-pheallaidh (Aberfeldy), 
Obair-readhain (Aberdeen), &c. Inverness- 
shire shows five or six of these ahers : Abertarf , 
or Mac Vurich's old Gaelic Obair-thairbh, so 
named from the Tarf or " Bull " river; Aber- 
arder (Laggan and Daviot), Gaelic, Obair-ardair, 
seemingly ' ' high-water ' ' ; Aberchalder (Glen- 
garry), where Calder appears, a name coimmon 
in Pictland. It first applied to water, the root 
is cal, sound, and the rest seems pure termina- 
tion -ent and -ar, the former a participial suffix, 
the latter an agent one. The name is un- 
doubtedly Pictish. With it may be compared 
the Gaulish river names Calarona, Callus, 
and Calla . The fourth name is Abriachan : 


in 1239 this was Abiihacyn, and in 1334 
Aberbreachy. Seemingly the streamlet entering 
Loch-Ness here must once have been called the 
Briachan; the curtailed phonetics reminds us of 
Arbroath from Aberbrothock. Abernethy, a 
name repeated in Fifeshire, is in Gaelic Obair- 
neithich, in 1239 Abyrnithy; the river is the 
Neithich. This has been equated with the Nith 
of Southern Scotland, which Ptolemy records as 
the Novios or ' ' Fresh ' ' (nuadh) stream, Welsh 
newydd. This would make the Pictish phonetics 
exceedingly Welsh and somewhat modern; but 
it is the best derivation offered. 

Two other words come to Gaelic from the 
Pictish, and are included in the ordinary voca- 
bulary. These are ' dul ' or ' dail,' ' a plain of 
fallow land, especially by a river-side,' and 
' preas,' ' a bush,' but in place-names, ' a 
brake.' The word ' dul ' or ' dail ' is exceed- 
ingly common as a prefix; as a suffix it shows the 
genitive ' dalach,' both in ordinary speech and 
places called Ballindalloch. The word does not 
appear in Irish, ancient or modern; but it is 
clearly allied to the similarly used word of similar 
meaning, W. ' dol,' pi. ' dolydd,' Corn, and 
Bret. ' dol.' Many place-names in Wales and 
Cornwall bear this prefix. The Perthshire parish 
name Dull, G. Dul, bears it in its naked sim- 
plicity, and the form ' dul ' is the usual one along 
the Great Glen, especially in Glen-Urquhart and 


Glen-Moriston. The modern spelling, however, 
is almost always ' Dal-' in these last cases. The 
Wardlaw MS. (17th century) always writes 
' Dul-', however. The root seems to be ' dul,' 
and therefore not allied to Eng. ' dale ' or Norse 
' dalr ' : but it is likelv alhed to the root ' dul/ 
bloom, as in Gaehc ' duilleag.' The word 
* preas ' is not common in place-names ; in the 
county we have it in Preas-mucrach (Badenoch), 
' Pig-brake place.' The Welsh word allied is 
■ prys,' brake, evidently alhed to the W. ' perth,' 
brake, whence the names Perth, Logie-Pert, Lar- 
bert, Partick, &c. The root, which is ' qr,' is 
that of G. ' crann,' W. ' pren.' 

Pictish influence may be seen in the common 
use of names rare or practically non-existent in 
Irish: monadh, hill, as in Monadh-liath ; blar, a 
plot, free space of ground — Blairour, ' Dun- 
plain ' (Lochaber), fi/dr-na-leine (1545) at 
the upper end of Loch Lochy; allt, a burn, 
Aldourie, from the " Dourag " burn, while 
Dourag itself is from dohhar, water; beinn, a 
hill, Irish heann, not much used in Irish place- 
names as compared to Gaelic ' beinn ' or ' ben ' ; 
cam, a hill, cairn, which Welsh also is fond of 
for names of hills, though not used in Ireland 
similarly — Cairn-gorm, Gealcharn, and others 
very numerous; coire, a corry or kettle — 
Corry Mhadagain, the " doggie's corry," a 
use of ' coire ' " scarcely known in Ireland " 


(Eeeves); srath, a strath, also a common 
Welsh and rare Irish word. The word 
which shows most departure from Gaelic use is 
both, a house, but used in Pictland for haile. It 
finds an especial development in Inverness 
county, particularly along the valley of the Great 
Glen — Bunachton, for Baile-Nechtain or Nec- 
tan's baile; Bochrubm, from old cruibin, a paw, a 
back-bent hill; Boleskine, in 1227 Buleske, from 
hoth-fhlescdin, " town of the withes," from 
flesc, a rod; Bolin (Glengarry), "flax-town"; 
and Bohuntin (Lochaber), where hunndainn 
stands for conntainn, a confluence. 

The use of 'rat,' apparently for 'rath,' a 
' fortified residence ' originally, in Strathspey 
and Badenoch, has also to be noted. The Welsh 
has the word ' rhath,' a clearing or open space, 
which seems to be the same word, and which 
Professor Rhys regards as borrowed from Gaelic. 
The exact extent of the use of ' rat ' in Pictland 
has not yet been considered, but on the analogy 
of Eothiemurchus, we might claim all the names 
in Rothie-, as Rothiemay. Raith in Fife, which 
certainly looks like the form that Pictish ' rat ' 
would assume, is claimed for Scotch ' wreath,' a 
pen, as are the several other names of like form. 
The matter is considered furthei' on under 

The first writer who gives any name bearing 
on Inverness-shire is Tacitus, who mentions the 


Caledonians, and the geographers represent 
them as extending into our county. Despite 
some difficulties in the classical form of the name 
Caledonia on the score of its phonetics not 
according with the root — that given being cald, 
the root of coille, English holt, nevertheless the 
nameDun/ceW and its Gaelic Dun-chailleann seem 
amply to prove that the classic Caledonia means 
really as the poet said, " land of the woods " — 
the Caledonians being the " Woodlanders."* 
Tacitus also records another famous name, 
Graupius, which has been misused in MSS., and 
appears most often as Grampius, whence comes 
the popular form Grampian. Tacitus meant 
some hill or hillock near Blairgowrie, but niedi- 
a3val imagination could fancy that nothing less 
could do justice to this great battle than the Gram- 
pian hills as a background and place of retreat. 
The root of Graupius is ' grup ' or, rather, 
' gruq,' and means 'hooked,' much as some 
hills are called 'sockach,' snouted. Ptolemy, 
the Geographer of 120 a.d., mentions the 
Vacomagi as the tribe inhabiting the " laigh " 

* Dr Stokes seiparatea the old Gaelic Cailleu or Caldeu from 
tha Classical Cakdonius, with its loug- e bet-ypeu / ana d; 
and the Welsh forms old and new (Celidon, C-elyddon; are 
certainly derived from the classical form, while the English 
form Dun-keld sliows the Welsh phonetics. The question is 
v-liether the cla&iieal form represents the real original; if so 
the roc'ts of Caillen and of Caledonia are not the eame. 


of Moray; the name divides as Vaco-Magi, the 
latter part being magh, a plain, the whole seem- 
ingly ' ' Dwellers on the plain . ' ' The name is lost . 
His name for Spey is Tvesis, which seems to have 
been an attempt at pronouncing Pictish initial 
sp, which in old Gadelic would be sqv, and in 
Welsh chiD — a troublesome sound. Dr Whitley 
Stokes explains Spey as Pictish, from the root 
sqe, as in sgeith, vomit, the Scotch spate, Welsh 
chwyd. The name appears to mean the 
" spatey, vomiting river," and it has the reputa- 
tion of being the swiftest of our large rivers. The 
Spean, on these terms, Vv^ould stand for Spesona, 
another stem from the same root. The Varar 
Estuary of Ptolemy answers to the Beauly Firth, 
and the River Farrar ideally suits the phonetics. 
The root may be var, crooKed. The Island 
Sketis, or better Skitis, which Ptolemy places 
about 70 miles north-east of Cape Orkas (Dunnet- 
Head), is probably the Isle of Skye misplaced, a 
view which commends itself to Muller, Thomas, 
and Stokes. The latter says that it is " the 
wing-shaped Island of Skye; Norse, Skidh; Irish, 
Scii (dat. case, date 700 in ' Annals of Ulster '), 
Adamnan, Scia; gen. Sceth (date 667 in ' Annals 
of Ulster '), Scith (Tigernach, 668) ; means wing, 
Ir. Sciath, Sciathan." Dr Stokes' derivation is 
the one usually accepted; the Norse Skidh, which 
is possibly influenced by ' ' folk-etymology, ' ' 
means a " log," " firewood," " tablet," and is 


allied to another Gaelic word sgiath, a shield. 
It is interesting to note that the Dean of Lismore 
refers to the island as " Clar Skeith " — the 
Board of Skith, thus showing that the Norse 
name of the island was remembered and trans- 
lated by Clar. More modern bards have used 
the expression Clar Sgith in regard to Skye. 
Thus Rory Mac Vurich in his elegy on Macleod 
(published in 1776) says : — 

" Dh' fhalbh mo lathaichean eibhinn 
O'n threio- sibh Clar Sgithe." 

In another on John, Sir Rory's son : — 

" 'S e 'n Clar Sgith an Clar raibh sgith." 

The earliest charter and record forms of the 
name Skye are Skey (1292), Sky (1336), and Ski 
in the " Manx Chronicle." Adamnan's ' Scia ' 
shows no trace of ' th.' The root is Celtic 
Ski, cut, slice, and the whole means the '"'in- 
dented isle." The root Ski is still the basis of 
Gaelic sgiath and Norse Skidh. 

Ptolemy's tribes in ancient " North Argyle " 
were the Creones, Cerones, and Carnonacae. 
The roots cer, ere, car, are here much to the 
front, and the root generally means " broken, 
rough." Carnonacae especially recalls cam, a 
cairn, a favourite name in the district as Ckrn, 
Carnan, and Carnach; to which may be added 
the Carron, the ' rough ' river, *Carsona. 


The title Hebrides, as applied to the Western 
Isles, appears first in Hector Boece's " History 
of Scotland." It is a copyist's blunder for the 
classical Hebudes or Haebudes, the name given 
by Phny to a group of the Western Isles, 30 ui 
number, he says. Ptolemy calls the Western 
Isles the Eboudae or Ebudae, five in number, of 
which two are named Ebuda. This made some 
writers attempt to identify the two ' ' Uists ' ' with 
the two Eboudae, but the phonetical difficulties 
here are too great; besides, the name Uist is, as 
Piofessor Munch said, simply the Norse word 
i-vist, a habitation. It has lately been conjec- 
tured that Ebouda stands for the Greek article 
(' e ' or '« '), plus Bouda or Boudda, or later 
B<')dda, and is really the old Pictish name of Bute. 
This would give that island name the meaning of 
" Victoria Isle." 

Adamnan, Abbot of lona, who died in 704, has 
left us in his " ' Life of St Columba ' ' the most 
important document that we possess bearing on 
the ancient history of our country. He has 
recorded seven or eight names belonging to 
Inverness County. Passing over his Dorsum 
Britanniae or Drum-Alban, which means the 
watershed of Argyle and Perth, continued north- 
wards also past the Great Glen, we have the 
names Nesa or Ness, Scia or Skye, Egea or Eigg, 

Airchartdan or Urquhart, Artdamuirchol or Ard- 



njimurchan, Aporicuvi Stagnum or Lochaber, 
and, lastly, the river whose Latin name is Nigra 
Dea (Black Goddess) in Lochaber. The river 
Ness is mentioned four times, three times as 
Nesa and once (in the genitive case) as Nisae. 
We learn also a lesson in topography from 
Adamnan — '" a 'cute ould observer," as an 
L'ishman would call him — Loch-Ness he calls the 
' ' Lake of the River Ness ' ' ; and it is almost 
invariably true, however large the loch or small 
the river, that the loch is named after the river 
which drains it. In addition to this, the river 
also names the glen through w^hich it flows; and 
we shall instantly find that the proud Ben Nevis 
i? named after, the humble nymph who once in 
pagan Pictish days ruled over the destinies of the 
Nevis stream. The name Ness is, of course, 
Pictish; and we need not look at modern Gaelic 
as exactly possessing the name in this form. We 
must have recourse to roots : Nesa, of Adamnan, 
points to Celtic Nesta and a root ned, which we 
find means "water," "wet," German netzerXy 
to wet, nass, wet, Sanskrit nadi, river. In old 
Greece there was the river Neda and in Thracia 
the Nestos, which is practically the "Ness." 
But we may go farther; in Ireland they had a 
heroic personage called Ness, mother of the 
famous demi-god king Conchobar Mac Nessa, 
who was, as can be seen, metronymically named. 
There are indications in the legends that Ness 


was really a river goddess of pagan Ulster; and, 
if so, we may regard the Pictish " Nessa " or 
' ' Ness ' ' as either the same goddess or her Celtic 
cousin. The Celts were great worshippers of 
rivers and wells. Gildas before 600 thus refers 
to the native worship of the early Britons .- 
'■ Nor will I invoke the name of the mountams 
themselves and the hills, or the rivers, to which 
the blind people then paid divine honour. ' ' One 
text represents Gildas as including the fountanis 
in the above enumeration, and we have m 
Ausonius (circum 380 a.d.), the Gaulish poet, 
an invocation to " Divona, fons addite 
divis," that is, " Divona, fountain dedicated to 
the Gods"; for the name meant "Goddess,"' 
and is the same as appears in the Ptolemaeic name 
for Aberdeen — Devana, which is still the Gaelic 
name of the river Don (Dian or Deathan), and 
which still abides in the -deen of Aberdeen. The 
river name Dee also means " Goddess " ; and we 
see from Adamnan that a river in Lochaber was 
called Nigra Dea or Black Goddess. Adamnan 
also mentions as in or on the Dorsum Britanniae 
the Lake of Loch-dae, and it has been well con- 
jectured that Loch-dae is the Gaehc or Pictish of 
Nigra Dea, for loch means "dark" and dae 
means " Goddess." In short, the river meant 
is the Lochy in Lochaber. There are at least 
four other rivers of this name : Lochay, entering 
the west end of Lochtay; Lochy in Glenorchy, 


entering the Orchy above Dalmally; Lochy, or 
Burn of Brown, which acts for a short distance 
as the boundary of Abernethy parish and Inver- 
ness county, and which joins the Avon at Inver- 
lochy near Kirkmichael ; and Lochy with Glen- 
Lochy at the head of Glenshee. 

We may, however, suspect more river names 
to have been '' Goddess " river names. This is 
undoubtedly the case with the " Earns," of 
which we have at least three or four : the Perth- 
shire Earn, the Inverness-shire Find-horn, or 
White Earn, and the Banffshire Deveron, or 
Doveran (oldest charter form Duff-hern) or Black 
Earn; and there is the Earn of Auldearn. The 
Earn of Strathdearn is called in Gaelic Eire, and 
its genitive is Eireann, the same in pronunciation 
as the name for Ireland, and it is the same as the 
name Erin of Ireland. Eire was one of the last 
Tuatha-de-Danann queens of Ireland, to which 
she left her name ; she was, in short, one of the 
last pagan female deities worshipped in Ireland. 
Ptolemy calls Ireland " Ivernia," and the Celtic 
form of the name is restored as ' ' Iverjo, ' ' or, 
possibly, a pre-Celtic Piverio (stem Piverion), 
which has been equated with the Greek land- 
name of Pieria, famed as the haunt of the muses. 
The root, in that case, would mean " rich, fat," 
and would scarcely apply to a river name. 
Adamnan's Evernihs, for " Irish," makes the 
whole matter doubtful, and at present we must 


confess ourselves beaten to explain the name 
" Eire " or " Eireann " — " another injustice to 
Ould Ireland'?"* I am inclined to include with 
these Goddess names also the name Nevis, the 
local Gaehc of which is Nihheis. This points to 
an early Pictish form — Nehestis or Nehesta, the 
latter possibly. The root neh or nebh is also 
connected with clouds and water, and gives us 
the classical idea of Nymph, root nbh — the 
fairies of Greece and Rome . The nymph Nebesla, 
then, gave her name to, or found her name 
in, the River Nevis, and gave her name to Glen- 
Nevis, and it again to the famous Ben, which 
again renders Inverness-shire unique, not merely 
among Scottish but among British counties, by 
having as one of its glories the highest hill in 
Britain. Loch Nevis also lends proof to the 
ai gument that Nevis really denotes water origin- 
ally. There was a river in ancient Spain called 
the Nehis, now Neyva, which may also show the 

Before leaving the river Ness and the other 
" Goddess " rivers of the district, I have to 

*Tli6 root 'pi' means "fat" and "drink," "water," 
" flow," aud is no doubt the ultimate root of these ' erin ' 
names, a stem ' pi-vo ' intervening, which is found in th« 
Gaelic name of lona, that is I, older Eo, li, Hii, from nom. 
' Piva,' loc. ' Pivi.' The rivers Esk, Ptolemy's Iska, are from 
*pid-ska, root pid, pi-d, spring, well, Grk. pid-ax, fountain. 
So likely Islay and Isla are from *pi-la. 


explain that there is another and more popuUii, 
possibly more poetic, derivation of the name 
Ness than the one I have offered. Once upon a 
time, the story goes, the Great Glen which now 
lies under the waters of Loch iSless was a beauti- 
ful valley, filled with people and plenty. In the 
bottom of the vale was a spring of magic virtue, 
but there was a geas or taboo connected there- 
with. Whenever the stone on the well was 
removed and the w'ater drawn, the stone had 
immediately to be replaced or else somethmg 
dreadful was to take place. One day a woman 
came to the well, leaving her child playing on 
her hut floor ; but while at the well she heard her 
child scream as if it had fallen into the fire. She 
rushed to the house to save her child, and forgot 
to replace the stone over the well. i'he w^ell 
overflowed at once, and soon filled the long 
valley. The people escaped to the hills and 
filled the air with lamentations, crying, ' Tlia 
loch nis ann; tha loch nis ann " — there is a lake 
there now^. The lake remained, and from that 
agonised cry is still known as Loch-Nis. 

Four other names in Adamnan still remani 
for us briefly to discuss — Egea, Aporicum, Art- 
damuirchol, and Airchartdan. His "Egea" 
Insula is the island of Eigg, the g of which we 
should expect to be aspirated nowadays, but 
here, as in the Ptolemaic Ebouda for Bute, and 
Adru for Ben Edair (Howth), the double sound 


of the consonant is not brought out in the old 
spelhng. Egea is for Eggea, and now it is in 
GaeKc Eige, old Gaelic genitive Ega or Eca. It 
is glossed or explained in a media3val MS. as 
" fons " or fountain, but the name seems to be 
the modern Gaelic eag, a notch. The island is 
notched, and so appears in approaching it. 
There is another " Egg " island off Glenelg, like 
in appearance. The Aporicum Stagnum or 
Stagnum Aporum — that is, the " Aporic lake " 
or ' ' lake of Apors ' ' — it is twice mentioned — 
is, of course, Lochaber. It is usual to regard 
the aber here as the Pictish prefix denoting 
" confluence," and, no doubt, " Loch of the 
Confluence ' ' of the Lochy with the Linne Dhubh 
(Black " Pool or Sea-loch ") — Loch Linnhe — is 
possible; but the Gaelic aher, a marsh, seems 
really to be the origin of the name, especially in 
view of Adamnan's plural Aporum or Ahers. 
" Loch of the Marshes," therefore, is the mean- 
ing of Lochaber. Artdamuirchol or Artdaib 
Muirchol is described as a " rough and stoney 
district"; it is known still as the Garbh- 
chriochan, and in the old charters we saw it was 
called Garmoran or Garbh-Morvern or ' ' Rough 
Morvern " — Morvern itself being in older Gaelic 
'Na Morbhairne " (genitive). In 1475 the 
records spell the name as " Morvarne " ; it can- 
not be Mor Earraiyin (Great Portion), as often 
explained, or Mor Bheannaibh; it is rather like 


Mor-hhearnaibh, " Great Gaps or Hill-passes." 
Coming back to Artdamiiirchol, the predecessor 
of Ardnamvrchane (1515), or now Ardria- 
murchan, we can easily divide the word into 
arda or ardaihh (accusative and locative plural 
of ard, high, height), and muirchol. This last 
Bishop Reeves explained as " Sea-hazel." 
Muir, sea, undoubtedly forms part of the word. 
There is no personal name of the form Mur-chol; 
so that Dr Reeves is probably right in hii 
" hazel " derivation. Lastly, we have Adam- 
nan's Airchartdan, which, of course, is Glen- 
Urquhart, the older " Wrchoden," and the 
modern " Urchadainn." There is an Urquhart 
in Cromarty, and another in Moray. The name 
is a compound : Air-card-an, the tirst element 
being the prefix air, on, beside. The second 
part, card or cardin appears in the oft-repeated 
Kincardine. It is clearly Pictish, and as Welsh 
cardd (older card) means " brake," we may 
take it that the Pictish means " wood, forest, or 
brake." Urquhart, therefore, means " Wood- 
side," as Kincardine means " Woodend." Cf. 
Welsh name Argoed, for ' ar-coed,' ' At Wood.' 
The word ' cardden ' is also found in Drum- 
chardine, older Drumcharding (1514), the for- 
mer name of Lentran . 

Let us now glance at the county from an 
ecclesiastical standpoint. There are thirty-five 
parishes in Inverness-shire, some of which it 


shares with its neighbours. Inverness town is 
in the territorial parish of Inverness and Bona; 
Bona refers especially to the Dochfour end of the 
parish, and is supposed to mean the ferry there 
crossing the Ness, called " Ban-ath " or 
" White-ford." In 1233 the parochial name 
was spelt Baneth, and two hundred years later 
Bonacht (for Bonath). The prefix cill, the 
locative of ceall, a church, appears in only four 
of the parishes, though it is otherwise common. 
Kil in Scotland almost invariably prefixes a 
saint's name ; it is the ceall of some saint. There 
are two or three exceptions, and the first on our 
hst is one of them : Kilmallie, Kilmalyn in 1296, 
Kilmale, 1532, means the church of Maillie, but 
there is no saint of that name, and it cannot be, 
as is often supposed, a pet corruption of Mairi or 
Mary. All cilJs dedicated to St Mary are Kil- 
moires or Kilmuirs, Moire being the real old 
Gaelic for St Mary, the name Mairi being of late 
Scoto-French origin. In Kilmallie parish is the 
river Mailhe and Invermailhe; we have also 
Kilmaly (1536), or Culmaly (1512), and Cul- 
mahn (1471) as the old name of Golspie parish; 
the stream at Golspie appears to have no name 
save Golspie Burn, so that it may have been 
called Maillie. There is a Dalmally in Glen- 
orchay, with an Allt-Maluidh running through 
it. There is Polmaly (' mailidh ') in Glen- 
Urquhart, with Allt-Phuill running into it, which 


must have been Allt-maly. Mailidh is a stream 
name; in Ireland Maiili is a personal name; but 
further than this I cannot go at present. Killin 
in Stratherrick, on Lochtayside, and at the upper 
end of Garve, means "White-church" {cill- 
fhinn), and is not, therefore, named after any 
saint any more than Kilmallie. In regard to the 
northern Killin there is the proverb — 

" Cill-Fhinn, Cill-Dumn 
'S Cill-Donainn— 
Na tri Cilltean is sine an Albainn." 

Kilvaxter, in Kilmuir of Skye, means the cill of 
Baxter, which got its name from the trade 
of somebody connected with it and the 
monastery of Monkstadt. Kilmore in Sleat 
means the Cella Magna or Great Church; 
there is a Kilmore in Glen-Urquhart. Kil- 
monivaig, Kilmanawik (1449), is the church 
of St Mo-naomhoc or " my saint " Naomhan 
Kilmoi'ack, Kilmorok (1437), seems dedicated to 
a St Moroc; the name has long puzzled ecclesi- 
astical students, but the form Mawarrock, a 
saint's name connected with Lecropt parish, at 
once suggests Mo-Bharroc, and we get the well- 
known St Barr or Barre, more fully Barr-finn or 
" White-head." There were several saints of 
the name, as also the name Finnbarr, the same 
name reversed, which was also curtailed to Barr, 
Findan, and Munn (Mo-Fhindu). The St Barr 


ol" Barra Isle was Finnbarr, whose day was on the 
25th September. Moroc's day was the 8th 
November. Kihiiuh', in Skye, means St Mary's 
Church, but the original name was Kilmaluok 
(1538) — Moluoc's or Lughaidh's Church, a 
favourite saint. Kiltarlity was in 1234 Kyl- 
talargy, in 1280 Keltalargyn; the saint is a 
Pictish one — Talorgan, " Fair-browed one." 

We have already discussed, in other connec- 
tions, Abernethy, Ardnamurchan, Boleskine, 
and Abertarff, Cawdor (under the name Aber- 
chalder, Cawdor being Caldor in 1394), Petty, 
Uist, Barra (that is Barr's ey or isle, mixed 
Norse and Gaelic), and Urquhart. Ardersier is 
in its oldest form Ardrosser (1226); it seems to 
mean Ard-rois-ear, " East-point-height," as 
against Ros-marky opposite it. The present 
pronunciation is Ard-na(n)-saor, ' Carpenters' 
Point ' ; but ' saothair,' a promontory or passage 
covered at high water, has been suggested. This 
word is common on the West Coast. Taking the 
Skye parishes together, we find Bracadale spelt 
much the same in 1498— BracadoU; the Gaehc 
is Bracadal; the name contains the common 
term breac or brae, slope, almost the same force 
as Gaelic sliahh, and it comes from the Norse 
brekka, a slope, EngHsh brink. Sleat, in 1389 
and 1401 Slate, comes from the Norse sletta, a 
plain, sUttr, level. It is the only decently level 
part of Skye. Strath is a curtailment of Strath- 


ordail ; it is a hybrid of Gaelic Sratli and Sword- 
dale or " Sward-dale," both Norse elements, 
usually Suardal in pronunciation. It is a very 
common name, this Swordale. Duirinish, in 
1498 Dyurenes, stands for Norse " Deer's ness 
or head." It is the same as Durness in Suther- 
land. Snizort is Snesfard in 1501 ; it possibly 
stands for Norse Snaesfjord or " snow-lirth." 
Portree doubtless gets its name of " King's 
Port " from James V.'s punitive visit to the Isles 
in 1540. 

Alvie parish, about 1350 Alveth and Aiway, 
presents a well-known name, which appears else- 
where as Alva, Alvah, Alves, and Alyth, which, 
save Alves, show an old form Alveth. It seems 
a Pictish stem alvo, an extension of the root al, 
rock. Daviot is another old word evidentlv 
Pictish, for its old form Deveth (1206-33) is 
clearly the same as the British tribal name 
Demetse of South Wales, now Dyfed. The root 
is dem, sure, strong, Gaelic deimhin. Croy 
and Dalcross formed an old parish. The former 
is from the adjective ' cruaidh,' hard. Dalcross 
is a corruption for what Shaw gives as Dealg-an- 
Ross or Dalginross, a name which appears in 
Athole and Strathearn. It means ' spit of the 
ridge or promontory,' for ' ros ' can be used 
inland, as in Abernethy^Euigh-da-ros, ' Shiel 
of the two points.' Dunlichity or Flichity 
is an alternate name for the parish; this 


is Flechate in 1560, and comes from flichead, 
moisture, a derivative from fliuch, wet. 
Dores, about 1350 Durrys, is in Gaelic Durus; 
this word meant in the old language " a gloomy 
wood " (duhhras), an epithet that would well 
suit the Inverness-shire Dores, if only the 
phonetics were more satisfactory. The name is 
Pictish — its termination (' -as ') favours this 
idea, and hence the root is ' dur,' strong — ' a 
strong hold,' it seems to mean. It has also been 
taken to mean ' dorus,' a door or opening; the 
roots in any case are the same. There is a 
Durris in Banchory parish. Duthil, about 1230 
Dothol, has been explained by Lachlan Shaw, 
the historian, as the tuaitheal or north-side of 
Creag-an-fhithich, while the Deshar or deiseil is 
on the south side. This also is the local deriva- 
tion, and it seems right enough. Glenelg, Glen- 
helk in 1282, means " noble glen," or properly 
the " glen of the noble {elg) river." The root 
elg is also in Elgin. Kingussie, Kinguscy 
(1103-11), is in Gaelic Cinn-ghii^ithsaich, 
"Head of the fir-forest"; cinn, or kin, as a 
prefix, is the locative of ceann. Kirkhill, a 
modern name, comprises the old parishes of 
Wardlaw (Wardelaw in 1203-24, an English 
name, meaning " Beacon-hill ") and Farnua 
(Ferneway in 1238). The latter name means 
the " place of alders " in Gaelic, and Shaw, who 
so explains it, adds that alders " abound there," 


which they have done till lately. Laggan is for 
Lagan-Choinnich or " St Cainneach's hollow," 
and in the old records it appears as Logynkenny 
(1239). The church was then up at the end of 
Loch Laggan. Moy is the locative of mayh, 
plain, and Dalarossie is in Gaelic Dail-Fhear- 
ghuis, the Dulergusy of 1224-42, the " dale of 
St Fergus," to whom the chapel there was dedi- 
cated. Rothiemurchus is in modern Gaelic 
Rat-a-mhurchais, which in 1226 is just the same, 
Eatemorchus, beside Rathmorcus. The prefix 
rat is a common one, confined, however, to Pict- 
land; it is an extension of rath, an enclosure or 
farm building, but whether the termination is 
due to Pictish influence or not can hardly be 
said; for in several cases d ends local suffixes, 
both in Ireland and Scotland (Irish kealid from 
caol, and croaghat from criiach; Scotch Bialaid 
in Badenoch, from hial, mouth). In fact riit 
takes the place of rath in Pictland ; and beside it 
we may no doubt place raig or rathaig as in 
Eaigmore and Raigbeg of Strathdearn, although 
the old forms show here an internal r : Ravoch- 
more; also Kil-ravock, which is now pronounced 
Kill-ra'ag. The main body of the word Rothie- 
murchus seems a personal name, possibly 
Muirgus, " Sea-choice," aUied to Fergus and 
Murchadh. The local derivation here is Rat- 
mhoir-ghiuthais, ' Rath of the big fir(s),' and is 


not to be despised on the score of phonetics, and 
certainly not as to the facts. 

The island parishes, besides Skye, comprise 
the Small Isles and the Outer Hebrides. Only 
Eigg now remains to Inverness-shire. Muck 
(Eilean-nam-muc or " Pig Isle "), Canna (Por- 
poise Isle, old Gaehc cana, porpoise), and Rum 
(origin unknown) belong now to Argyle. St 
Becan, from ' bee,' ' beag,' little, seems to have 
died in Eum (gen. Euimm) in 676, if we can 
judge what the Irish annals and mar- 
tvrologies sav correctlv. Eig2 has been 
already considered. So, too, have the 
LHsts and Barra. Harris was in 1546 Hary, 
1546 Harige; Dean Munro (1549) calls it " the 
Harrey." The Gaehc is Na h-Earra, which 
gave the Enghsh form ' ' the Merries ' ' and 
Harris or " the Harris." There is Harris in 
Rum and Islay, Herries in Dumfries, and Harray 
ill Orkney. It is usual to explain Na h-Earra 
as " the heights," and both in Harris and in 
Islay this admirably suits, but the Norse words, 
whence the name undoubtedly comes, cannot be 
easily fitted in. The Norse for " high " is hdr, 
plural havir, especially the comparative haerri, 
higher (" The Higher Ground " as compared to 
low-lying Lewis). 

The Church has supplied many other than 
purely parish names. Saints' names, generally 
wit the prefix cill, are abundant, and saints' weUs, 


as well as saints' isles, are common. St Columba 
is first favourite, something like a score of places 
being connected with his name in such forms as 
Cill-cholumchille (Kil-columkill) or Cill-choluim, 
Tobair-Cholumchille, and Eilean-Cholumchille ; 
and Portree bay was named after him originally. 
The next in importance of dedication is the 
Virgin Mary; Kilmuir or Kilmory are the usual 
forms in English of the name. There are 
two in Ardnamurchan, Kilmory and Kil- 
vorie, Kilmuir in North Uist, and Kilmuir 
in Skye as a parish, and in Duirinish, with 
several other places. St Bridget, the "Mary 
of the Gael," has two or three Kilbrides in the 
county — as in Strath, South Uist, and Harris. 
St Maolrubha, older Maelruba, appears in place 
names as Molruy, Morruy, and Maree (as in 
Loch-Maree). His centre in Scotland is Apple- 
cross; here he died in 721. He seems 
to have been a favourite in Skye ; there is 
Kilmaree in Strath, and Cill-ashik was of old 
Askimolruy or " Maelruba 's Ferry " ; Kilmolruy 
in Bracadale, and Ardmaree in Berneray. 
In Skye also Moluag or St Lughaidh has 
some dedications — Kilmaluock in Trottarness 
and in Raasay; there was a croft Mo-luag at 
Chapelpark, near Kingussie, whence the latter 
name. St Comgan is celebrated in Ardna- 
murchan and Glenelg — Kilchoan; and he was 
the special patron of the old Glengarrv family. 


St Cuimine the Fair, the Tth century biographer 
of Cohimba, seems to have been celebrated at 
Glenelg, Kh'kton (Kilchuimen, 1640). But we 
have his name certainly in Cill-chuimen of Fort- 
Augustus. St Donnan gave Kildonnan to Eigg 
and South Uist. The Pictish saint Drostan, who 
is misrepresented as a pupil of St Columba's, was 
patron of Alvie; his chapel is still seen in ruins 
nt Dunachton, and there is, or was, in Glen- 
Urquhart a croft named after him — Croit-mo- 
chrostan; and seemingly the patronymic M'Eostie 
(Perthshire) comes from Drostan under Lowland 
influence. Another Pictish saint was Kessoc, 
whose name at least is borne by the ferry of 
Kessock (Kessok, 1437). The name Kessoc or 
Kessan is from ' ces,' meaning ' spear ' in Gaelic, 
but what it meant in Pictish it is impossible to 
say. Tarlagan, the Pict, had a ' kil ' on the 
north of Portree bay, besides being the patron 
saint of Kiltarlity (Ceilltarraglan). Adamnan ap- 
pears rarely; Tom-eunan of Insh is named after 
him, and a croft of his existed in Glen-Urquhai't. 
Such names as Kilpheder, Kilmartin, Kilaulay 
(Olave), Kilchalman, Kilcrist (now Cill-chro, or 
'' pen kirk," in Gaelic, in Strath), Pitkerrald 
(Cyrill), and Kilmichael in Glen-Urquhart, 
Killianan (Finan) in Glengarry, Ardnamurchan 
and Abiiachan, and others can only be men- 



A most interesting ecclesiastical name ii 
Annaid; it occurs very often in Inverness county, 
from Killegray of Harris to Groam of Beauly. 
Achnahannet is common, and there are Teampall 
na h-Annaid, Clach na h-Annaid, and Tobair na 
h-Annaid. It means in old Gaelic a patron 
saint's church; it is rare, however, in Ireland, 
and seems in Scotland to denote the locale of the 
pioneer anchorites' cells, that is, their clachans 
and Httle oratories, often away in a diseart (Lat. 
Desertuvi) or desert (island or remote place). 
The name Clachan is common on the West Coast 
and in the Isles; it means, firstly, the monk's or 
anchorite's bee-hive stone cell — built where 
wood and wattle were scarce, so that on the 
eastern mainland there are no clachans. The 
word developed into the meaning of oratory or 
kirk, and, from the cluster of clachans making a 
monastic community, into " village," which is 
its only meaning in the Lowlands. There are 
three in Kilmuir (Skye), for example; one at 
least in N. Uist, which is countenanced by Kallin 
or Ceallan (Kirkie) and Kirkibost (" Kirkton ") 
there. Reilig is now an old Gaelic word for 
church-yard, and from Lat. reliquiae; it appears 
in the Aird and near Beauly as Euilick. Team- 
jpull and Seipeal (Chapel) give many names : 
Tigh-an-Teampuill or Temple-House in Glen- 
Urquhart, and Pairc-an-t-seipeil (Chapel-park) 
in Badenoch, for example. The chm'ch officials, 


too, have naturally left their mark : Balnespick 
is Bishop's-ton; Paible is from the Norse Papyli 
or Papa-byU, " Pope or Priest's town," a Gaehc 
Bail'-an-t-sagairt, and Papay is " Priest's Isle" ; 
Mugstad or Monkstead of Skye is the half Norse 
representative of Bal-vanich in Benbecula, which 
is half Gaehc {manach, monk, from Lat. 
monachus). In the same island is Nunton or 
Ballenagailleich (1549). There is no Appm in 
Inverness-shire — Abbacy or Abbey-land, but 
there is "A' Mhanachainn," the Monkery, the 
the Gaelic name for Beauly, itself from the Lat. 
Bellus Locus or " Beautiful Place," a name no 
doubt bestowed on it- — and rightly — by the early 
13th century monks. 

We shall now notice the District names not 
already considered, as we have considered Loch- 
aber, Morvern, Strathdearn, &c. The Aird 
explains itself; it is the high ground of Kirkhill 
and Kiltarlity. Glenmoriston is a difficult name ; 
the river, of course, gives the name, and it is 
usually explained as for Mor-easan, "river of 
great water-falls." It is Pictish, no doubt, and 
points to a Celtic ^M6r-est-ona. Stratherrick, the 
older Stratharkok and Stratharkeg, comes from 
the river Farigag, which means " lower ravine " 
river (Gaelic far, below, and gag, cleft). Far is 
a common prefix in northern Pictland — Farleitir 
("lower slope"), Farraline ("lower linn"), 
Farr ("lower place"), &c. Strathnairn derives 


its name from the Nairn Eiver; this river name 
is Pictish, hkely old Naverna, the same root and 
partial stem as we have in the Naver of Suther- 
land, Ptolemy's Nabaros. The root is uav or 
snav, flow, swim, Gaelic ,snamh; and we may 
compare the Welsh Nevern as a parallel form to 
Nairn. Badenoch is the Gaelic Bkideanach; the 
root is haide, submerged, from hath, drov.n. In 
Ireland there is Bauttogh in Galway, " a marshy 
place," and the river Bauteoge, running through 
.Avampy ground. Passing over Lochaber as 
already discussed, we come to the ancient lorrl- 
ship of Garmoran, the Clanranald land, bounded 
on the south by Loch Shiel and on the north by 
Loch Hourn, as the poet says in the Dean of 
Lismore's Book (1512) — 

Leggit derri di vurn 
eddir selli is sowyrrni 

— " An end of merriment between Shiel and 
Hourn." Adamnan's Sale is the above Shiel, 
but the Sorn is a later name given by the Gael, 
who had by the time they reached it adopted the 
Latin ' furnus,' whence ' sorn,' a furnace, un- 
doubtedly comes. Loch Hourn is ' Furnace 
Lake ' — Lochshuirn, which may be compared 
with the Lochalsh name Coire na Soma, the one 
a mascuhne, the other a feminine genitive, both 
genders being shown in the the early language, 
as is not uncommon in the case of a borrowed 


word. The lordship of Garmoran, to which 
Skene devoted an extraordinary chapter in his 
" Highlanders of Scotland," under the fancy 
that it was an earldom, and about which he is 
silent in "Celtic Scotland," comprised Moydart, 
Morar, and Knoydart. The name, spelt in 1343 
Garmorwarne, means ' Eough Morvern,' and 
Morvern means ' Great Passes ' — Mor-bhearna ; 
the modern Gaelic has adopted the name Garbh- 
chriochan, or ' Rough-bounds,' instead. The 
Morvern furthers south may be regarded as 
adjacent, and perhaps part of the same name; 
if not, then it also is bisected well enough by its 
own ' beam ' or pass of Lochs Tacnis, Loch 
Arienas, and, we may add. Loch Aline, with 
their respective streams, to entitle it to a 
separate but singular Mor-bhearn. M'Vurich 
calls it in the gen. sing. fem. ' Na Morbhairne ' ; 
the oldest charter spelling is Morvern as now 
(1390), and Morvarne (1475). The name Moy- 
dart, G. Miiideard, was spelt Mudeworth in 
1343, Modoworth in 1372, and Mudewort in 
1373. The name is difficult as to derivation; it 
is Norse by its ending ' -ard,' ' -ort,' which is 
for ' fjord.' Like Knoydart and Sunart, it likely 
comes from a personal name, here Mundi, and 
for the phonetics compare the island names 
Gometray and Hermitra, from Godmund and 
Hermund, and the personal name Tormoid from 
Thormund. Better still is the Throiid of 


Trotternish for comparison. Siinart, in 
1372 Swynwort, and m 1392 Swynawort, is 
Sveinn's fjord; while Knoydart (Cnudeworth in 
1343) stands for Knut's or Canute's fjord. 
Arisaig, in 1309 Aryssayk, is the Norse dros-vik, 
the bay of the river mouth {dros, river -mouth, 
whence Aros, the place name). Morar was in 
1343 Morware, Mordhowor, 1517 Moroyn, Mac 
Vurich's old Gaelic Moiroin, which last points to 
Mor-shron or " Great nose " (promontory) as 
the meaning of the word; but Morar or Morwar 
stands for Mor-bharr,' Great-point.' Glengarry 
takes its name from the river Gareth (about 1309). 
There is another Garry in Perth, and the Yarrow 
is the same name, while allied by root are the 
English rivers " Yair " and " Yare " (Yar- 
mouth), and also the French " Garonne," classic 
Garumna. The root is garu, or Gaelic garhh, 
rough. In Skye we have Trotternish, Water- 
nish, and Minginish districts. Trotternish is in 
1549 both Trouteruesse and Tronternesse, either 
with a u or with n in the main syllable. Mac 
Vurich (17th century) gives the then Gaelic as 
" Trontarnis " ; it stands for Norse " Thrond- 
arnes " or " Thrond's Headland." Waternish 
is the Icelandic " Vatnsness" or " Water-ness." 
Minginish— Myngnes in 1498, Mygnes in 1511, 
and Myngynnes in 1549— contains the prefixed 
element ming, which appears in the island names 
Mingulay and Mingay, and Mingarry, where in 


every case the Gaelic has no ' ng ' sound at all. 
Mingarry is Mioghairidh (Mewar, 1493, and 
Meary, 1505, but Mengarie, 1496). The word 
here prefixed seems to be ' mikil,' ' great,* 
whose accusative is ' mikinn,' ' mikla,' ' mikit ' 
in the three genders. Hence Minginish means 
Eudha-Mor of GaeHc, which it is. 

The Norsemen, who held the Isles for some 
450 years, have left a deeper impress on the 
place-names there than the Gael. Of the name* 
usually printed on maps, in directories, or in the 
Valuation Eolls for the Outer Hebrides, four are 
Norse to the Gaelic one; that is, the proportion 
is four-fifths Norse and one-fifth Gaelic. In 
Skye the proportion is not so heavily against 
Gaehc; practically the two languages are equal. 
Of the names on the Valuation Roll, 60 per cent, 
are Norse as against 40 per cent, that are Gaehc. 
The coast-line of Garmoran is also considerably 
Norse, though nothing like the proportion in 
Skye; and as we go inland the Norse names get 
fewer. There are no Norse names in Lochaber; 
so we may conjecture that that district was free 
of the Norse yoke. Norse names abound in 
Easter as well as in Wester Ross, and they can 
be traced south to the Beauly valley, where we 
have Eskidale (" Ash-dale ") and Tarradale on 
the Beauly River. Further south we do not find 
any trace of the Norse power in place names; 
nor is it likely that they ever had any conquest 


or sway south of Beauly, despite their own asser- 
tions, in their sagas, that they possessed also 
Moray. The Norse power in Scothand at its 
strongest extended over Caithness, Sutherland, 
Ross, Argyle, and Galloway, with, of course, th« 
Weste]-n Isles. This was about 980 to 1050. 
Gaelic slowly regained its hold in the Isles after 
the rise of Somerled and the other patriarchs of 
the Clan Donald in the latter part of the 12th 
century; but Gaelic in its re-conquest left the 
Norse nomenclature of the country practically 

The most prominent Norse words borrowed 
are those for island {ey), hill ifjciU), vik 
or -aig, bay, nes or nish, headland, dalr 
or -dale, a vale, a dale; fjordhr, sea-loch 
or firth (fjord), or -ord, -ard, and the various 
words for township, farm or settlement (setr, 
stadr, holstadr and bol or -bo). The termina- 
tion -ay and -a of the island names is the Norse 
ey, isle. Beginning with the isles about Harris, 
we have Berneray, or " Bjorn's Isle" — Bjorn 
either meaning ' ' bear ' ' or being a personal 
name, which last it likely is. Fladda, so com- 
monly repeated, means " flat isle "; Soay, also 
repeated often, is for Saudha-ey or " Sheep- 
isle " ; Isay, " Ice-isle " ; Taransay, St Taran'» 
Isle; Ensay, " meadow (engi) isle " ; Killegray, 
"Kellach's Isle," the Kellach bemg the Irish 
•^Cellach" or "Kelly" (Warrior), borrowed 


early by tlie Norse, and now known in the name 
MacKillaig; Lingay, "Heath Isle''; Scalpay, 
" Shallop^" or " Ship Isle "; Eossay, " Horse 
Isle"; Eriskay, "Eric's Isle." Oransay and 
Orasay, of which names there are a great number 
of isles, is from orjiri, ebb or shallow, and means 
that the island is one at full tide only; Pabbay, 
"Pope or Priest's Isle"; Sandray is "Sand 
Isle ' ' ; Benbecula is only partly Norse ; the 
Gaelic is Beinn-a-bhaodhla, and really means 
" Height of the Ford," from Gaelic faodhail, 
" a ford," itself borrowed from the Norse vadill, 
"a shallow or ford." Easay or Earsay 
(Eairsay, 1526, Easay and Eaarsay in 1549) 
seems to be ' Ear-ass-ey,' ' Eoe-ridge-isle.' 

The hills in the isles generally end in -val. 
This is the Norse fjall, fell or hill. The name 
Eoine-val is common ; this is Hraun-fell, a rocky- 
faced hill; the island Eona is also from hraun, 
"rocky-surfaced isle." Horne-val is "horn- 
fell "; Helaval is " flagstone fell "; and so on. 
Layaval in South Uist, and Laiaval in North Uist, 
may be equated with Ben Loyal in Sutherland; 
perhaps for ' Leidhfjall,' ' levy or slogan hill.' 
Mount Hecla in Mingulay has the same name as 
the famous burning mountain in Iceland, which 
means ' hooded shroud.' Blavein in Skye is for 
Bid-fell, 'Blue-fell.' 

The sea-lochs in -ord, -ard, -art are too 
numerous even to make a selection from; and 


the same may be said of the nesses or headlands 
(Norse nes). I must pass over also the town- 
ships with their hols, hosts, and sias. An odd 
change is undergone by holmr, an islet (in a bay 
or river), a holm; this may appear either as ter- 
minal -am, or -mid, or -lum. We have Heista- 
mul and Hestam, both from hestr, horse; tiie 
famous Eilean Beagram is probably Bekra- 
holmr, "Ram-holm"; Lamalum is "Lamb- 
holm," and Sodhulum is from saudhr, sheep. 
Airnemul is Erne-holm — " Eagle-holm." Liani- 
mul no doubt means "flax-holm." Os means 
river-mouth, oyce " ; we have it in the Skye 
Ose and Glen-ose, and in Aros. Hoe and Toe 
are not uncommon, and we have Howmore in 3. 
Uist; this is Norse haugr, burial mound, howe. 
Torgabost shows horgr, a heathen place of 
worship, and also Horogh (Castlebay). 

There is a marked difference between the 
Island and West Coast topography and the 
eastern mainland in the common names of hills, 
dales, lochs, and rivers; in the west we have 
cleit, stac, sgiirr, sgeir, and gil, all Norse; in the 
east cam, meall, creag, monadh, and gleann. 
In the east coire, srath, sliahh, as against the 
terminal dal and hreac and gil of the Isles. 
Then the absence of terms for wood is most 
marked in the west, sco, terminal for skogr, a 
shaw, appearing only in Skye, as Birkisco, 
Grasgo, &c. In the east, wood is very common 


in the nomenclature. The bird names also differ 
much, even when not Norse, from the Gaelic 
Mainland. We have orri, N. moorfowl, also a 
nickname, in Oreval, hills in Harris and Uist; 
mar, sea-mew, in Maraig, ' Sea-mew bay ' ; dm, 
eagle, in Arnamul, ' Eagle-head ' (Mingulay), 
and Arnaval (Skye); krdka, crow, Crakavick, 
* Crow-wick ' (Uist) ; hrafn or hramn, raven, 
Eamasaig, ' Eaven-bay ' (cf. Eamsay, Eamsey); 
and Geirum, ' Auk-holm ' (Barra). 

The mainland haile, farm or township, is 
often represented in the Inverness-shire isles by 
Norse setr, a stead, shieling. The latter name 
appears alone as Seadair (Gaelic) or Shader 
(English) in Bernera and ^ Skye. Uigshader 
means ' Ox-ton ' (compare Uisgeval and Uisg- 
neval, hills) ; Eoishader, ' Horse-ton ' ; Mari- 
shader ' Mare-ton ' ; Herishader, ' Lord's-ton ' ; 
Sulishader, ' Pillar-ton ' or ' Solan-goose-ton ' — 
it is not far inland — all in Skye ; which, however, 
prefers host (N. holstadhr), as Husabost, 
'House-stead'; Eabost (' Eidh ' or isthmus?); 
Colbost (pronounced Cyalabost), ' Keel-ton ' ; 
Heribost, ' Lord's-ton ' ; Orbost, ' Orri's-ton ' ; 
Breabost, ' Broad-ton ' ; Skeabost, ' Skidhi's- 
ton,' as in Skibo (old Scythebol); Carbost, 
' Kari's-ton.' The Norse gardr, a garth or 
house and yard, which appears elsewhere on 
Norse ground, is represented in the Western 
Isles and Mainland by its diminutive gerdhi, 


which has been adopted into Gaelic as gearrnidh, 
the land between machair and moor. It is com- 
mon in place-names in its Gaelic use — Gearadu, 
' B-lack-garth,' in N. Uist; Geary (Duirinish); 
Garry more (Bracadale), Garrafad (Kilmuir), and 
Gairidh-Ghlumaig (Kilmuir). Terminally it is 
garrij, and is very extensively used with Norse 
names — Osmigarry, from Osmund; Calligarry, 
from Kah; Grimagarry, from Grimm; Shageary, 
' Sea-garth ' (Sagerry, 1541); Flodigarry, 'Float 
or Fleet garth ' (though Gaelic has long o) ; Big- 
gary, ' Barley ' ; Mugeary, ' Monk's garth ' (?); 
and Mosgaraidh, ' Moss ' — all in Skye. In N. 
Uist there are Hougheary (ho we), and Trums- 
garry (Thrum's); in Benbecula, Creagarry may 
be Gaelic, as may be Crogarry there, though 
kro may be Norse borrowed from Gaelic (a pen) ; 
Mingarry (Benbecula) is ' mickle-garth.' In S. 
Uist appears Stelligarry, the first portion of 
which is pronounced ' staol,' and is found in 
Stulay isle; it is Norse, pointing to steil, steyl, 
stadhil or stagil, but these forms are either non- 
existent or cannot be used in place-names, save 
the last, as in Stagley, 'rock-isle.' Seemingly 
we have here a corruption of the proper name 
Stulli or Sturla. The Norse has borrowed be- 
sides kro the imporant word airigli, shieling, 
originally as aerg or erg, as in Asgrims-aergin, 
in the Orkney Saga, where it is explained that 
erg is Gaelic for setr. Asgrims-erg now appears 


as Askarry, even Assary (Caithness), where we 
have also Halsary (Hall), Dorrery, Shurrery 
(Shureval, ' Pig-hill,' in S. Uist), etc. In Duir- 
inish we find Soarary, ' Sheep shiel ' ; in Ardna- 
murchan, Smirisary, 'Butter shiel,' and Brunary 
(Brunnary, 1498), an ' Airigh an tobair ' ; in 
Glenelg, Beolary and Skiary ; in N. Uist, Obisary, 
' Bay or Hope ' ; Aiilasary, ' Olaf 's ' ; Eisary, 
' Copse-wood ' ; Dusary, Vanisary, and Horisary ; 
in S. Uist, Vaccasary and Trasaiy (Thrasi); and 

Some of the more interesting land and farm 
names may be glanced at. The Norse ounce 
and penny lands — especially the latter — have left 
their mark. The tirung or ounce-land is equated 
with the Mainland davoch or dock, four plough- 
gates, whose fourth is the common name Kerrow 
(ceathramh, fourth). The Norse for this last 
phonetically was fjordlimigr, fourthing or farth- 
ing, which appears in the place-name Feoirlig, the 
phonetics being the same as for hirlinn, a galley 
(N. byrdhingr). It meant ' farthing land.' The 
ung was old Gaelic, and existed in Ung-an-ab, 
the abbot's ounce-land, in N. Uist in 1561. The 
pennyland gives many names : Pein-chorran 
(Portree), from corran, point, the masculine 
form of corrag. This corran is a very common 
name in the Isles, and appears as Corran simply 
several times, as at Ballachulish. The usual ex- 
planation of ' bay ' is absurdly wrong, therefore, 


fiom corran, a sickle, supposed metaphorically 
to mean 'bay,' which it does not. Of course 
these ' corrans ' often guard sickle-shaped bays, 
and hence the mistake. Other penny-lands are 
— Penifiller, ' Fiddler's '; Pensoraig, 'Primrose' 
(?orN. 'Saur-vik," Mud-bay'); Pein-more (big) ; 
Peiness (waterfall); Peinaha; Peinlich; Leiphen 
(half-penny); and Pein-gown (smith) — all in 
Skye. Peinavaila is the romantic form which 
' Peighinn-a'-bhaile ' takes in Benbecula. Pen- 
inerin in S. Uist stands for * Peighinn an aor- 
ainn' — where mass was said. In Pictland 
davoch or dock is the commonest land-measure : 
Dochgarroch, ' D. of the rough-land ' ; Doch- 
four, of which presently; and Lettoch, near 
Beauly, is ' Half-davoch,' like the Aberdeen- 
shire Haddo and Haddoch. The terminal ele- 
ment -fur enters largely into the names of Pict- 
land — Balfour, Inchfur, Dalfour, Dochfour, Pit- 
fur (very common), Tillifour and Tillifourie 
(Tough), and Trinafour (Perthshire). The form 
with / is clearly an aspirated p ; the word is pur, 
which seems to exist in diminutive form in Purin 
(Fife), older Pourane, Porin (G. Porainn) in 
Sirathconan, and Powrie near Forfar. The 
Book of Deer has the aspirated Fiirene, repre- 
sented now by Pitfour in Deer. The p proves 
the word to be Pictish ; and it is possible that the 
root is par, as in Welsh paivr, pasture, Breton 
peur. The ultimate root is qer, as in preas, 


crann, and perhaps craohh. In Inverness we 
have Dochfour, Dochgarroch, and Delfour. 

The words gart, corn, goirtean, cornfield, 
alhed to Enghsh garden and Norse gardhr, ap- 
pear in Boat of Garten and minor places. Cluny 
is a very common name ; the Gaelic is Cluanaigh, 
a locative of cluanach, meadowy place, from 
cluan, a mead. In Badenoch the nom. or ace. 
is found in A' Chluanach, west of Kincraig. 
Longart, a shieling, camp, is now obsolete, save 
in place-names; it is met with inDail-an-longairt, 
Coire-an-Longairt, and Badenlongart (1773, 
Gaick) — all in Badenoch. The old word was 
longphort, ' ship-port,' or harbour, encamp- 
ment, which, with a dialect pronunciation of 
long as ' low,' gives luchairt, a palace. Tarbert 
means isthmus, from tar, across, and root her, 
bring, bear. Drummond presents the full stem 
of druiyn, back (dromann, dromand), and does 
not stand,' as usually said, for Druim-fhinn, 
white ridge, still less for Fionn's ridge. Strath- 
glass presents the old word glais, stream, which 
we have in Inveruglas, the confluence of the 
Duglas or Dark-stream (now nameless) ; this is 
also found in Southern Scotland, and has given 
the famous family name. The word leacainn, 
a cheek, hill face or side, gives Leachkin, at 
Inverness, and elsewhere, generally with an 
epithet. The diminutive sldhean, a fairy knoll, 
gives Baihntian and many names else ; the simple 


sldh appears in Ben Tee, of Glengany, and is 
found elsewhere for conical hills, as hi Schie- 
hallion, ' Hill of the Caledonians,' with which 
the name Dunkeld and Rohallion, near Duukeld, 
are to be compared. The lairig is giveri in the 
dictionaries as a " plain, hill, sloping hill," 
somewhat contradictory meanings; but the real 
meaning is found in the place-names, and that 
meaning is ' pass.' In Old Irish we have Idarc, 
a fork or ' gobhal.' Finnlarig, both in Duthil 
and at Killin, means ' Fair Pass,' as Eev. J. 
Maclean, Grandtully, etymologises the Perth- 
shire name. In Eothiemurchus we have Larach- 
grue or Lairig-dhru, probably the pass of Druie 
river (root dru, flow, as in Gaulish Druentia), 
which the Ordnance Map, with its wonted per- 
versity, names Lairg Gruamach. The place- 
name Elrick is common in the county, and theie 
must be over a hundred such in Scotland ; it is 
from the obsolete eileirig, locative of eileireag, 
which meant the cul-de-sac bounded by fallen 
trees and other obsti'uctions into which the deer 
v;ere driven, and one side of which was formed 
of a hill, on the face of which the hunters took 
their place and shot the deer. These hills and 
places are called Elrick, Eldrick, Elrig, and 
Ulrig ; ' eileir ' is given in the dictionaries as a 
* deer path,' no doubt from the root elu in eilid, 
hind. It is sometimes explained as iolairig, a 
knoll on which eagles rested, which is not likely. 


The ' bordlands ' of the royal and other castles 
appear in Gaelic as horlum, whence Borlum, 
near Fort-Augustus, also the old name for Ness 
Castle, whence the famous and notorious Borlum 
family got its name. There is Borlum in Skye, 
and elsewhere. 

We will finally consider some interesting indi- 
vidual names, and begin with the furthest west, 
which is St Kilda. This name is one of those 
known as ' ghost names ' — a geographer's 
blunder. In Gaelic the island is called Irt or 
7ori, which means in old Gaelic ' death ' ; it is 
likely that the ancient Celts fancied this sunset 
isle to be the gate to their earthly paradise, the 
Land-under-the-waves, over the brink of the 
western sea. The Dutch map-makers of the 
17th century are responsible for St Kilda or 
Kilder. There were some wells near the village 
famous for their virtues — Tobar-nam-buadh, 
and there was a Tobar-Kilda among them — one 
or all of them retaining the Norse name for well, 
which is kelda, corrupted into St Kilder's Well 
in the 17th century. Kelda is known in the 
North of England on Norse ground as kild, as in 
Kiidwick, Kiiham (Domesday Chillum), and 
Halikeld, 'Holy-well.' The well-names got 
mixed with the true name of the island on the 
maps. The Dutch were active herring fishers in 
the western seas in 'he 17th century, and to 



them we owe more curiosities than St Kilda — 
doubtless the Minch is due to them, the Gaehc of 
which is A' Mhaoil, the Moyle, also the old Irish 
name for the sea between the ' Maoil ' of Kintyre 
and Ireland. 

Rodel, o long, stands for Norse Red-dale, 
from the colour of the soil. 

Lee, in N. Uist, Ben Lee, Skye, N. hlidh, 

Lochmaddy, from viadadh, a shellfish there. 

Heisker, Hellisker, 1644, N. 'Rocky skerry.' 
Munro in 1649 calls it Helskyr na gaillon (nuns). 

Stoney-bridge, in S. Uist, G. Staoni-bris, is 
for N. Stein-brekka, 'stone-slope.' 

Boisdale, N. Bugis-dalr, ' SKght bay dale.' 

Dorlin, Ardnamurchan, G. doiWin^,' isthmus. 

Glenfinnan, G. Gleann-Fionain, named after 
St Finan, Ellan-Finan having been the old name 
of Ardnamurchan parish. St Finan lived in St 
Columba's time, is called of ' Swords in Lem- 
ster,' and was latterly a leper, taking the 
infection for penance. His name appears in 
Abriachan and Glengarry in Killianan. He is 
not to be confused with St Finnan (short i, from 
finn, white); as the following triplet on the last 
Glengarry shows, the quantity of the i is long : — 

'S ann 'na lai^he !n Gill Fhionain 
Dh' fhag sinn biatach an fhiona, 
Lkmh a b' urrainn a dhioladh. 


Inveraros, in Eaasay, is a good case of 
hybrid; for dros is the Norse for inver. 

Point of Ayre, in Eaasay, is derived from 
eyrr, a gravelly beach, connected in Britain with 
headlands; we have it in Snizort as Eyre (Ire, 
1630), and Ken-sal-eyre or Kinsale (sea-end) of 
Eyre. There is a Point of Ayre on the north- 
east coast of Man; and we may perhaps conjoin 
the Heads of Ayr in the county of that name, 
and perhaps the county name. 

Idrigill, which appears twice as a promontory 
in Skye, with Udrigle in Gairloch, stands for 
Ytri-kolir, 'Further or Outer Hill.' It is not 
connected with gil, a ravine. 

Bealach Colluscard (Kilmuir) is interesting 
again as showing tautology, for Collu-scard 
means Pass of the Hill (kollr), N. skardhr. It 
is again repeated in Bealach na Sgairde in 
Portree, with somewhat ugly emphasis. 

Armadale is Norse, meaning ' bay-dale.' 

Skulamus (Strath) seems to be for Skull's 
moss, while Strolamus must be for Sturli's moss 
(for u as 6, compare Knoydart, which has a 
liquid also). 

Broadford is a modern name, not Norse. 

Tahsker, G. Tallasgar, N. T-hallr-sker, 
' Sloping rock.' 

Eist (Duirinish), a Chersonese, is from hestr, 
horse, that is, 'horse-shaped.' Otherwise, as 
in Eilean Heist, it really means ' Horse'-isle. 


Greshornisli (Duirinish), pronounced Grls- 
innis now usually, is for Grice or Pig Ness. 

Eigj^ (Snizort) and Digg (G. Dig) are respec- 
tively from Norse hryggr, ridge, and dik, a ditch. 
Duntulm is the dim of the holmr, islet. 
Staflin, 'The Staff,' from N. stafr, a staff, 
applied to basaltic and other pillared rocks, as m 
Staff a (basalt isle) and Dunstafnage (Dun- 
staffynch, 1309), Dun-stafa-nes. 

Loch Arkaig (Lochaber), river Arkaig, from 
Celtic root arc, dark, W. erch, dusky; Loch 
Arklet, Stirling. 

Corpacli, ' place of bodies.' Here, it is said, 
the bodies carried to Zona for burial rested to 
await sailing. 

Banavie, Banvy (1461); compare Banff, 
Bamff, also Banba, an old name for Ireland, 
from hanhh, a pig. For meaning, compare 
Mucrach and Muckerach (Kilmorack), Pres- 
Mucrach, mucrach meaning ' Place of Pigs.' 

Fersit, Farset (Bleau), from obsolete /earsaicf, 
sandbank at the mouth of a river, whence also 

Fassfern, G. Fasaidh-fearn, ' Abode or stead 
of the alders.' 

Glen-quoich, Glen of the Cuaich river, the 
river of cuachs or bends. It is a common river 

Loch Oich; Oich points to a Celtic Utaka, 
root lit, dread , * awesome.' 


Vinegar Hill, Gaick, is in Gaelic ' A Mhin 
Choiseachd,' the easy walking. The English is 
a fancy name. 

Ettridge is for Eadar-dha-eas, ' Between two 
falls.' NessintuUich, EssintuUich (1645), is for 
'- Water-fall of the hillock.' Phoines is for Fo 'n 
eas, ' Below the fall.' So with Phoineas in Kil- 

Coylum Bridge; Gaehc, Cuing' leum, ' Nar- 
row leap,' which it is. 

Achnacoichen (Rothiemurchus), ' Field of 
the Owls ' ; so in Lochaber — Achnacochine, in 
1509 Auchancheithin. 

Eothiemoon (Abernethy), G. Rat a' mhoin, 
'Rath or stead of the peat-moss. 

Pityoulish, in Abernethy, older Pitgaldish, is 
Pictish in prefix, root, and termination {-ais). 
The root word is geall, pronounced hke the word 
for ' promise.' It is found in many river names : 
Geldie Burn, running into Upper Dee; Aber- 
geldie; Innergeldie near Comrie; Innergelly in 
Fife (river Gelly); perhaps Lochgelly there; 
Glen-geoulhe near Cawdor; AUt Gheallaidh at 
Dalnacardoch and Knockando. The root is 
geld, as in Norse kelda, a well, Ger. quelle, 
already mentioned in connection with St Kilda. 
A shorter form of the root is found in G. geal, a 
leech, root gel, water. Compare Welsh Aber- 


Granish (Duthil), G. Greanais (Gren-), for 
older Granais, apparently from grain, abhor- 
rence; but likely Pictish, denoting ' rough place,' 
from the same root and stem. The place figures 
largely in Druid lore and writings on account of 
its stone circles, and is consequently called 
Grianais, ' Sun-place,' which does not agree 
with the modern pronunciation. 

Aviemore, G. Agaidh-mhor; there is also 
Avinlochan, the Avie of the loch. Gallovie, as 
in G. Gealagaidh and present Blairgie, waa 
writtetn in 1603 as Blairovey — both in Laggan. 
Ayaidh may be Pictish; compare Welsh ag, cleft, 
opening, Gaelic eag. 

Craigellachie, whence the war cry of the 
Grants, has its name from eileach, place of 
rocks, rock, old Gaelic ail, rock. It is a much 
be-bouldered and rock-ribbed bare hill. 

Morile (Strathdearn), G. Moir'l, seems to 
stand for a Pictish Mor-ialon, ' Large clearing,' 
Welsh ial, open space. Hence, too, Balmoral. 

Kyllachy, G. Coileachaigh, ' Place of moor 

The Cuigs of Strathdearn, or fifth parts, are 
famous : " Is fhe^rr aon choige' an Eireann na 
coig choige' an Strath-Eireann " — "Better is 
one fifth in Ireland than the five fifths in Strath- 
dearn." The Irish fifth is a province, such as 
Ulster. The ' Cuigs ' were — Ciiig-na-fionn- 


druinich (' Bronze Place,' perhaps a smith's 
place), Cuig-na(n)-scalan (tents or huts), Cuig- 
na-sith (fairy hill, near is the Sidh-bheinn, the 
Schiphein of the charters), Cuig-na-fearn 
(alders), and, likely, Ciiig-na-muille (mill). 

Scaniport, ' Cleft of the Ferry,' over the 

Foyers, old Foyer, for old Gaehc fothir, good 
land, evidently ' low-lying land,' as the land of 
Foyers along Lochness is. 

Allt-saidh (Glen-Urquhart), ' Burn of the 
hound (female).' 

Fort- Augustus, from William Augustus, Duke 
of Cumberland, so named by General Wade, 
circ. 1730. 

Fort- William, the fort built at Auchintore 
(Bleaching-field) for William of Orange; also 
Maryburgh for the village, from Mary, his con- 
sort; then Gordonsburgh, from the dukes of 
Gordon, who disliked ' Orange ' ; and Duncans- 
burgh, on the ' passing ' of the Gordons, from 
Sir Duncan Cameron of Fassfearn; and now 
finally settled as Fort- William. 

Fort-George, built in 1748, takes its name 
from the King. The original Fort-George was 
the Castle of Inverness. 

Essich, Essy in 1456, a locative of easach, 
water-fall stream, rapidly falling stream. The 
name exists in Strathbogie, Forfar, and Moray. 


Castle Heather presents an interesting ' ghost 
name.' Further back it was Castle-leather, 
older Lathir, and we find the Lordship of Leffare 
(1456) apphed to the district along the slope 
there. It comes from leathair, a side, found in 
Leathair nam Manach, at Beauly, ' Monks' Side 
of the Valley,' 'Monks' Hillside '—the Kil- 
morack district east of Breakachy Burn. In the 
west, we have An Leathair Mhorairneach and An 
Leathair Mhuileach — the coastland of Morvern 
and of Mull. 

Culloden, Cullodyn in 1238, present GaeUc 
Cuil-fhodair, 'Fodder-nook,' by popular etymo- 
logy. It really comes from lodan, a pool, and 
means ' Back of Pool,' or ' Nook of Pool.' As 
in many similar cases, there is quite a shower of 
* cuils ' near Culloden, going over the Nairn 
valley, ending with Cuil-chuinneig, ' Nook of the 
iwooden pail,' apparently. It was here that 
Prince Charles' staff was stationed before the 

Brochnain is for Bruach 'n-eidheinn, ' Ivy 

Tomnahuirich, GaeHc of 1690 Toim-m- 
hurich, ' Hillock of the Yew-wood.' The Ward- 
law MS. gives both Tomnihurich and Tom ni 
Fyrich. This last may account for the deriva- 
tion of the name from Tom-na-fiodhraich, fiodh- 
rach being alleged to mean 'wood' (A. Mac- 


kenzie in Inverness Field Club Trans., III., p. 


Erchless, a quoad-sacra parish, (H)erchely8 
in 1258, Ercles, 1403, Arcles, 1512, appears to 
stand for air-glais, On the Glass — the river Glass 
passes through the Mains of Erchless. Compare 
the neighbouring Urray from Air-rath, On-fort 
or Repaired Fort, and Urquhart and Urchany of 
Beauly and Nairn (air-canach). The Gaelic is 

Glen Affric takes its name, as does the loch, 
from the river Affric, which has the old female 
name Afric or Oirig (Euphemia), and which 
comes from ath-hreac, somewhat-speckled, from 
hreac, speckled, a trout. Here it was no doubt 
a water-nymph's name. 

Glen-Con vinth and Convent, which was an 
old parish, appears in old records as Conveth and 
Conway, and in Gaelic the name is Confhadh- 
aich, which, appHed to the river, means ' noisy, 
stormy,' from cvnfhadh, storm. 

Lovat, older Loveth, seems a Pictish word 
(root lu, stem lu-vo, mud) translated into Gaelic 
as A' Mhor'oich, the sea-side plain or swamp. 

Two districts of Inverness-shire have had 
their names discussed in detail, and both can be 
relied upon as much as any work done in this 
paper. The districts are Badenoch, which is 
considered in the Transactions of the Gaelic 


Society of Inverness, Vol. XVI., pp. 148-97, 
and Urquhart and Glenmoriston, the place-names 
of which are fully discussed in Dr Mackay's 
work, " Urquhart and Glenmoriston," pp. 5 TI- 



Mil; iTh 




Badenoch is one of the most interior districts 
of Scotland; it lies on the northern watershed 
of the mid Grampians, and the lofty ridge of 
the Monadhlia range forms its northern 
boundary, while its western border runs along 
the centre of the historic Drum-Alban. Even 
on its eastern side the mountains seem to have 
threatened to run a barrier across, for 
Craigellachie thrusts its huge nose forward into 
a valley already narrowed by the massive form 
of the Ord Bain and the range of hills behind 
it. This land of mountains is intersected by 
the river Spey, which runs midway between the 
two parallel ranges of the Grampians and the 


Monadhlia, taking its rise, however, at the ridge 
of Drum-Alban. Badenoch, as a habitable 
land, is the valley of the Spey and the glens that 
run off from it. The vast bulk of the district is 
simply mountain. 

In shape, the district of Badenoch is 
rectangular, with east-north-easterly trend, its 
length averaging about thirty-two miles, and its 
breadth some seventeen miles. Its length along 
the line of the Spey is thirty-six miles, the river 
itself flowing some 35 miles of the first part of 
its course through Badenoch. The area of 
Badenoch is, according to the Ordnance Survey, 
551 square miles, that is, close on three hundred 
and fifty-three thousand acres. The lowest 
level in the district is 700 feet; Kingussie, the 
" capital," is 740 feet above sea-level, and 
Loch Spey is 1142 feet. The highest peak is 
4149 feet high, a shoulder of the Braeriach 
ridge, which is itself outside Badenoch by about 
a mile, and Ben Macdui by two miles. Moun- 
tains and rivers, rugged rocks and narrow glens, 
with one large medial vallev fringed with culti- 
vation — that is Badenoch. It is still well 
wooded, though nothing to what it once must 
have been. The lower gi'ound at one time must 
have been completely covered by wood, which 
spread away into the vales and glens ; for we find 
on lofty plateaux and hill sides the marks of 
early cultivation, the ridges and the rigs or 


feannagan, showing that the lower ground was 
not very available for crops on account of the 
forest, which, moreover, was full of wild beasts, 
notably the wolf and the boar. Cultivation, 
therefore, ran mostly along the outer fringe of 
this huge wood, continually encroaching on it as 
generation succeeded generation. 

The bogs yield abundant remains of the once 
magnificent forest that covered hillside and glen, 
and the charred logs prove that lire was the 
chief agent of destruction. The tradition of the 
country has it that the wicked Queen Mary set 
fire to the old Badenoch forest. She felt 
offended at her husband's pride in the great 
forest — he had asked once on his home return 
how his forests were before he asked about her. 
So she came north, took her station on the top 
of Sron-na-Baruinn — the Queen's Ness — above 
Glenfeshie, and there gave orders to set the 
woods on fire. And her orders were obeved. 
The Badenoch forest was set burning, and the 
Queen, Nero-like, enjoyed the blaze from her 
point of vantage.^ But many glens and nooks 
escaped, and Eothiemurchus was left practically 
intact. The Sutherlandshire version of the 
story is different and more mythic. The King 
of Lochlann was envious of the great woods of 
Scotland; the pine forests especially roused his 
jealous ire. So he sent his muime — it must 

1 Queen Mary ravaged Huntly's lands, and burnt the woods. 


have been — a witch and a monster, whose name 
was Dubh-Ghiubhais, and she set the forests on 
fire in the north. She kept herself aloft among 
the clouds, and rained down fire on the woods, 
which burnt on with alarming rapidity. People 
tried to get at the witch, but she never showed 
herself, but kept herself enveloped in a cloud 
of smoke. When she had burned as far as 
Badenoch, a clever man of that district devised 
a plan for compassing her destruction. He 
gathered together cattle of all kinds and their 
young; then he separated the lambs from the 
sheep, the calves from the cows, and the young 
generally from their dams ; then such a noise of 
bleating, lowing, neighing, and general Babel 
arose to the heaven that Dubh-Ghiubhais popped 
her head out of the cloud to see what was wrong. 
This was the moment for action. The Badenoch 
man was ready for it ; he had his gun loaded with 
the orthodox sixpence; he fired, and down came 
the Dubh-Ghiubhais, a hfeless lump ! So a part 
of the great Caledonian forest was saved amons 
the Grampian hills. 

Modern Badenoch comprises the parishes of 
Laggan, Kingussie and Insh, and Alvie; but the 
old Lordship of Badenoch was too aristocratic 
to do without having a detached portion some- 
where else. Consequently we find that Kincar- 
dine parish, now part of Abernethy, was part 
of the Lordship of Badenoch even later than 


1606, Vv^lien Huntly excambed it with John of 
Freuchie for lands in Glenhvet. Kincardine 
was always included in the sixty davachs that 
made up the land of Badenoch. The Barony of 
Glencarnie in Duthil — from Aviemore to Garten 
and northward to Inverlaidnan — was seemingly 
attached to the Lordship of Badenoch for a time, 
and so were the davachs of Tullochgorum, Curr, 
and Clurie, further down the Spey, excambed by 
Huntly in 1491 with John of Freuchie. On the 
other hand, Eothiemurchus was never a part oi 
Badenoch, though some have maintained that ii 
was. . The six davachs of Hothiemurchus be- 
longed to the Bishops of Moray, and at times 
they feued the whole of Eothiemurchus to some 
powerful person, as to the Wolf of Badenoch iii 
1383, and to Alexander Keyr Mackintosh in 
1464, in whose family it was held till 1539, when 
it passed into the hands of the Gordons, and 
from them to the Grants. 

Badenoch does not appear in early Scottish 
history; till the 13th century, we never hear of 
it by name nor of anything that took place witli- 
in its confines. True, Skene, in his CeUir 
Scotland, definitely states that the battle of 
Monitcarno was fought here in 729. Thi.s 
battle took place between Angus, King of 
Fortrenn, and Nectan, the ex-king of the Picts, 
and in it the latter was defeated, and Angus 



shortly afterwards established himself on the 
Pictish throne. We are told that the scene of 
the battle was ' ' Monitcarno jiixta stagnum 
Loogdae " — Monadh-carnach by the side of 
Loch Loogdae. Adamnan also mentions Loch- 
dae, which Columba falls in with while going 
ijwer Drum Alban. Skene says that Loch Insh 
— the lake of the island — is a secondary name, 
and that it must have originally been called 
Lochdae, that the hills behind it enclose fthe 
valley of Glencarnie, and that Dunachton, by 
the side of Loch Insh, is named Nectan's fort 
af)ter King Nectan. Unfortunately this view is 
wrong, and Badenoch must give up any claim 
to be the scene of the battle of Monadh-carno ; 
Lochdae is now identified with Lochy, and Glen- 
carnie is in'Duthil. But Dunachton is certainly 
Nectan's fort; whether the Nectan meant was 
the celebrated Pictish King may be doubted. 
Curiously, local tradition holds strongly that a 
battle was fought by the side of Loch Insh, but 
the defeated leader was King Harold, whose 
grave is on the side of Craig Righ Harailt. 

From 729, we jump at once to 1229, exactly 
five hundred years, and about that date we find 
that Walter Cumyn is feudal proprietor of 
Padenoch, for he makes terms with the Bishop 
of Moray in regard to the church lands and to 
the " natives " or bondsmen in the district. It 
has been supposed that Walter Cumyn came into 


the possession of Badenoch by the forfeiture and 
death of Gillescop, a man who committed some 
atrocities in 1228— such as burning the 
(wooden) forts in the province of Moray, and 
setting fire to a large part of the town of Inver- 
ness. Wilham Cumyn, Earl of Buchan, the 
justiciar, was intrusted with the protection of 
Moray, and in 1229 Gillescop and his two sons 
were slain. Thereafter we find Walter Cumyn 
in possession of Badenoch and Kincardine, and 
it is a fair inference that Gillespie was his pre- 
decessor in the lordship of Badenoch. The 
Cummings were a Norman family; they came 
over with the Conqueror, and it is asserted that 
they were nearly related to him by marriage. 
In 1068, we hear of one of them being governor 
or earl of Northumberland, and the name is 
\3ommon in English charters of the 12th century, 
in the early part of which they appear in Scot- 
land; they were in great favour with the 
Normanising David, and with William after him, 
fining offices of chancellors and justiciars under 
them. William Cumyn, about the year 1210, 
married Marjory, heiress of the Earldom of 
Buchan, and thus became the successor of the 
eld Celtic Mormaers of that district under the 
title of Earl of Buchan. His son Walter 
obtained the lordship of Badenoch, as we saw, 
and, a year or two after, he became Earl of 
Menteith by marrying the heiress, the Countess 


of Menteith. He still kept the lands of 
Badenoch, for, in 1234, we find him, as Earl of 
Menteith, settling a quarrel with the Bishop of 
Moray over the Church lands of Kincardine. 
Walter was a potent factor in Scottish pohtics, 
and in the minority of Alexander III. acted 
patriotically ias leader against the pro-English 
party. He died in 1257 without issue. John 
Comyn, his nephew, son of Eichard, succeeded 
him in Badenoch; he was head of the whole 
family of Comyn, and possessed much property, 
though simply entitled Lord of Badenoch. The 
Comyns at that time were at the height of their 
power; they could muster at least two earls, the 
powerful Lord of Badenoch, and thirty belted 
knights. Comyn of Badenoch was a prince, 
though not in name, making treaties and kings. 
John Comyn, called the Eed, died in 1274, and 
was succeeded by his son, John Comyn, the 
Black, and in the troubles about the kingly 
succession, at the end of the century, he was 
known as John de Badenoch, senior, to distin- 
guish him from his son John, the Eed Comyn, 
the regent, Baliol's nephew, and claimant to 
the throne, whom Bruce killed under circum- 
stances of treachery at Dumfries, in 1306. 
Then followed the fall and forfeiture of the 
Comyns, and the lordship of Badenoch was 
given, about 1313 — included in the Earldom of 


Moray — to Thomas Randolph, Bruce' s right- 
hand friend. 

The Cummings have left an ill name behind 
them in Badenoch for rapacity and cruelty. 

Their treachery has passed into a proverb — 

" Fhad bhitheas craobh 's a choill 
Bithidh foill 'sna Cuiminich." 

Which is equally smart in its English form — 

' ' While in the wood there is a tree 
A Gumming will deceitful be." 

It is in connection with displacing the old pro- 
prietors — the Shaws and Mackintoshes — that (.he 
il] repute of the Cummings was really gained. 
Bu(, the particular cases which tradition remem- 
bers are mythical in the extreme; yet there is 
something in the traditions. There is a remem- 
brance that these Cummings were the first feudal 
lords of Badenoch; until their time the Gaehc 
Tuath that dwelt in Badenoch had lived under 
their old tribal customs, with their tbiseachs, 
their aires, and their saor and daor occupiers 
of land. The newcomers, with their charters, 
their titles, and their new exactions over and 
above the old Tuath tributes and dues, must 
have been first objects of wonder, and then of 
disgust. The authority which the Cummings 
exerted over the native inhabitants must often 
have been in abeyance, and their rents more a 
matter of name than reality. However, by 


making it the interest of the chiefs to side with 
them, and by granting them charters, these 
initial difficulties were got over in a century or 
two. It was under this feudahsing process that 
the system of clans, as now known, was 

Earl Randolph died in 1332, and his tvro 
sons were successively Earls of Moray, the 
second dying in 1346 without issue, when 
" Black Agnes," Countess of Dunbar, 
succeeded to the vast estates. The Earldcm of 
Moray, exclusive of Badenoch and Lochaber, 
was renewed to her son in 1372.^ Meanwhile, 
in 1371 Alexander Stewart, King Robert's son, 
was made Lord of Badenoch by his father, as 
also Earl of Buchan ; and in 1387 he became Earl 
of Ross through his marriage with the Countess 
Euphame. His power was therefore immense; 
he was the King's lieutenant in the North (locum 

1 Sir W. Fraser, in his " History of tlie Grants," says : 
— " After the forfeiture of the Comyns, Badenoch formed a 
part of the earldom of Mora}% conferred on Sir Thomas Ean- 
dolph. In 1338, however, it was held by the Earl of Eoss, 
and in 1372, while granting the Earldom of Moray to John 
Dunbar, King Robert II. specially excepted Lochaber and 
Badenoch." Sir W. Eraser's authority for saying that 
Badenoch was in the possession of the Earl of Ross must be 
the charter of 1338 granting Kiurara and Dalnavert to Mel- 
moran of Gleucharny; but a careful reading of that document 
shows that the Earl of Ross was not superior of Badenoch, 
for he speaks of the services due by him to the " Lord superior 
of Badenoch." Besides, in 1467, wh^n Huntly was Lord of 
Badenoch, we find the Earl of Ross still possessing lands 
there, viz., Invermarkie, which he gives to Cawdor as part of 
his daughter's dovirry. 


tenens in horealihus partihus regni) ; but such 
was the turbulence and ferocity of his character 
that he was called the "Wolf of Badenoch." 
He is still remembered in the traditions of the 
country as " Alastair Mor Mac an Eigh " — 
Alexander the Big, Son of the King — a title 
which is recorded also in Maurice Buchanan's 
writings (a.d. 1461, Book of Pluscarden), who 
says that the wild Scots (Scotis silvestrihus) 
called him " Ahtstar More Makin Ee." Natur- 
ally enough he gets confused with his famous 
namesake of Macedon, also Alastair Mor, but the 
more accurate of tradition-mongers differentiate 
them easily, for they call Alexander the Great 
" Alastair Uabh'rach, Mac Eigh Phihp" — Alex- 
ander the Proud, son of King Philip." This 
epithet of uahh'rach or uaihhreach appears as 
applied to Alexander the Great in that beautiful 
mediaeval Gaelic poem that begins — 

" Ceathrar do bhi air uaigh an fhir 
Feart Alaxandair Uaibhrigh : 
Eo chansat briathra cen bhreicc 
Os cionn na flatha 6 Fhinnghreicc." 

Translated — 

Four men were at a hero's grave — 

The tomb of Alexander the Proud; 

Words they spake without lies 

Over the chief from beauteous Greek-land.^ 

1 See " Dean of Lismore," p. 84; Ranald Macdonald's CSol- 
lection, p. 133, and Highland Monthly, II., p. 376. ("Ihe 
above is from a British Museum MS.). 


The Wolf of Badenoch's dealings with his 
inferiors in his lordship are not known ; but that 
he allowed lawlessness to abound may be in- 
ferred from the feuds that produced the Battle 
of Invernahavon (circ. 1386), and culminated 
in the remarkable conflict on the North Inch of 
Perth in 1396. We are not in much doubt as 
to his conduct morally and ecclesiastically. He 
had five natural-born sons— Alexander, Earl of 
Mar, Andrew, Walter, James, and Duncan — a 
regular Wolf's brood for sanguinary embroil- 
ments. He had a chronic quarrel with Alex- 
ander Bur, Bishop of Moray, w^hich culminated 
in the burning of Elgin Cathedral in 1390. But 
in nearly every case the Bishop, by the terrors 
of the Curse of Eome, gained his point. In 
1380, the Wolf cited the Bishop to appear be- 
fore him at the Standing Stones of the Eathe 
of Easter Kingussie (apud le standand slanys de 
le Eathe de Kyngucy estir) on the 10th October, 
to show his titles to the lands held in the Wolf's 
lordship of Badenoch, viz., the lands of Logach- 
nacheny (Laggan), Ardinche (Balnespick, &c.), 
Kingucy, the lands of the Chapels of Eate and 
Nachtan, Kyncardyn, and also Gartinengally. 
The Bishop protested, at a court held at Inver- 
ness, against the citation, and urged that the 
said lands were held of the King direct. But 
the Wolf held his court on the 10th October : the 
Bishop standing "extra curiam" — outside the 


couri, i.e., the Standing Stones — renewed his 
protest, but to no purpose. But upon the next 
day before dinner, and in the great chamber 
behind the hall in the Castle of Euthven, the 
Wolf annulled the proceedings of the previous 
day, and gave the rolls of Court to the Bishop's 
notary, who certified that he put them in a large 
fire lighted in the said chamber, which consumed 
them. In 1381, the Wolf quits claims 
on the above-mentioned church lands, but in 
1383 the Bishop granted him the vv^ide domain 
of Rothiemurchus — " Ratmorchus, viz., sex 
davatas terre quas habemus in Strathspe et le 
Badenach " — six davochs of land it was. The 
later quarrels of the Wolf and the Bishop are 
notorious in Scotch History : the Wolf seized 
the Bishop's lands, and was excommunicated, 
in return for which he burnt, in 1390, the towns 
of Forres and Elgin, with the Church of St Giles, 
the maison dieu, the Cathedral, and 18 houses 
of the canons. For this he had to do penance 
in the Blackfriar's Church at Perth. He died 
in 1394, and is buried in Dunkeld, where a hand- 
some tomb and effigy of him exist. 

As the Wolf left no legitimate issue, some 
think the Lordship of Badenoch at once reverted 
to the Crown, for we hear no more of it till it 
was granted to Huntly in 1451. On this point 
Sir W. Fraser says: — "The Lordship of 
Badenoch was bestowed by King Eobert II. upon 


his son, the ' Wolf of Badenoch,' in 1371, and 
should have reverted to the Crown on the Lord 
of Badenoch' s death in 1394. But there is no 
evidence in the Exchequer Eolls, or elsewhere, 
of any such reversion, and Badenoch seems to 
have been retained in possession by the Wolf of 
Badenoch's eldest son, who became Earl of Mar. 
. . . Alexander, Earl of Mar, and his father, 
were therefore the successors of the Comyns as 
Lords of Badenoch." 

The Lordship of Badenoch was finally 
granted to Alexander, Earl of Huntly, by James 
II., by charter dated 28th April, 1451, not in 
recompense for his services at the Battle of 
Brechin, as is generally stated, but upwards of a 
year before that event. The great family of 
Gordon and Huntly originally came from near 
the Borders. Tliev obtained their name of 
Gordon from the lands of Gordon, now a parish 
and village in the west of the Merse, S.W. 
Berwickshire. There, also, was the quondam 
hamlet of Huntly, a name now represented there 
only by the farm called Huntly wood. The 
parish gave the family name of Gordon, and the 
hamlet of Huntly gave the title of Earl or 
Marquess of Huntly. Sir Adam de Gordon was 
one of Bruce' s supporters, and after the for- 
feiture of the Earl of Athole he got the lordship 
of Strathbogie, with all its appurtenances, in 
Aberdeenshire and Banff. The direct male 


Gordon line ended with Sir Adam's great-grand- 
son and namesake, who fell at the battle of 
Homildon Hill in 1402, leaving a daughter 
Elizabeth, who married Alexander Seaton, 
second son of Sir W. Seaton of Winton. Her 
son Alexander assumed the name of Gordon, 
and was created Earl of Huntly in 1449. His 
son George was Lord Chancellor, founded 
Gordon Castle, and created the Priory of 
Kingussie (Shaw's Moray). The Gordons were 
so pre-eminent in Northern politics that their 
head was nicknamed " Cock of the North." 
In 1599, Huntly was created a Marquis, and in 
1684 the title was advanced to that of Duke Df 
Gordon. George, the fifth and last Duke of 
Gordon, died in 1836, when the property passed 
into the possession of the Duke of Eichmond and 
Lennox, as heir of entail, in whose person the 
title of Duke of Gordon was again revived ;n 
1786, the full title being now Duke of Richmond 
and Gordon. 

Save the Church lands, all the property in 
Badenoch belonged to Huntly either as superior 
or actual proprietor. The Earl of Ross 
possessed lands in Badenoch under the lord 
superior in 1338, which he granted to Malmoran 
of Glencarnie : the lands were Dalnavert and 
Kinrara, and the grant is confirmed about 1440, 
while in 1467 we find the Earl of Ross again 
granting the adjoining lands of Invermarkie to 


the Thane of Cawdor, in whose name they 
appear till the seventeenth century, when Inver- 
eshie gets possession of them. The Laird of 
C J rant, besides Delfour, which he had for three 
centuries, also held the Church lands of Laggan 
and Insh, that is, " Logane, Ardinche, Bally- 
naspy," as it is stated in 1541, and he is in 
possession of them for part of the seventeenth 
century. Mackintosh of Mackintosh has in feu 
from Huntly in the sixteenth century the lands 
of Benchar, Clune, Kincraig, and Dunachton, 
with Rait, Kinrara, and Dalnavert. The only 
other proprietor or feuar besides these existing 
in the 16th century seems to have been James 
Mackintosh of Cask. The Macphersons, for 
instance, including Andrew in Cluny, who 
signed for Huntly the " Clan Farsons Band." of 
1591, are all tenants merely. We are very 
fortunate in possessing the Huntly rental of 
Badenoch for the year 1603. Mackintosh ap- 
pears as feuar for the lands above mentioned, 
and there are tvN^o wadsetters — Cask and Strone, 
both Mackintoshes. The 17th century sees 
quite a revolution in landholding in Badenoch, 
for during its course Huntly has liberally granted 
feus, and the proprietors are accordingly very 
numerous. Besides Huntly, Mackintosh, and 
Grant of Grant, we find some twenty feus or 
estates possessed by Macphersons; there was a 
Macpherson of Ardbrylach, Balchroan, Benchar, 


(in) Blarach, Breakachie, Clune, Cluny, 
Corranach, Crathie, Dalraddy, Delfour, 
Etteridge, Gasklyne, Gellovie, Invereshie, 
Invernahaven (Inverallochie), Invertromie, 
Nuid, Phones, and Pitchirn. There vv^as a Mac- 
kintosh of Balnespick, Benchar, Delfour, Gask, 
Kinrara, Lynwilg, Rait and Strone — eight in all. 
Four other names appear once each besides 
these during the century — Maclean, Gordon of 
Buckie, Macqueen, and Macdonald. The total 
valuation of Badenoch in 1644 was £11,527 
Scots, in 1691 £6523, and in 1789 it was 
£7124, with only seven proprietors — Duke of 
Gordon, Mackintosh, Cluny, Invereshie, Belle- 
ville, Grant of Grant (Delfour), and Major 
Gordon (Invertromie). The " wee lairdies " of 
the previous two centuries were swallowed up 
in the estates of the first five of these big pro- 
prietors, who still hold large estates in 
Badenoch, the Duke of Gordon being repre- 
sented by the Duke of Richmond since 1836. 
Only one or two other proprietors on any large 
scale have come in since — Baillie of Dochfour, 
Su' John Ramsden, and, we may add, Macpher- 
son of Glentruim. The valuation roll for 1889- 
90 shows a rental of £36,165 lis 7d sterling. 



In the above section we discussed the 
pohtical history of Badenoch, under the title of 
the "Lordship of Badenoch," and in this 
section we intend to deal with the history of the 
native population of that district. Badenoch 
was the principal seat of the famous and power- 
ful Clan Chattan. The territory held by this 
clan, hov/ever, was far from being confined to 
Badenoch ; for at the acme of their power in the 
15th century. Clan Chattan stretched across mid 
Inverness-shire, almost from sea to sea — from 
the Inverness Firth to near the end of Loch-eil, 
that is, from Petty right along through Strath- 
nairn, Strathdearn, and Badenoch to Brae- 
Lochaber, with a large overflow through Rothie- 
murchus into Braemar, which was the seat of 
the Jarquharsons, who are descendants of the 
Shaws or Mackintoshes of Rothiemurchus. The 
Clan Chattan were the inhabitants of this vast 
extent of territory, but the ownership or 
si^eriority of the land was not theirs or their 
chiefs', and the leading landlords they had to 
deal with were the two powerful Earls of Huntly 
and Moray. From them, as superiors. Mackin- 
tosh, chief of Clan Chattan, held stretches of 
land here and there over the area populated by 
the clan, and his tribesmen were tacksmen or 
feu-holders of the rest, as the case might be,, 


under Moray or Huntly. It was rather an 
anomalous position for a great Highland chief, 
and one often difficult to maintain. Major 
(1621) describes the position, territorially and 
otherwise, of the Clans Chattan and Cameron in 
words which may be thus translated : — ' ' These 
tribes are kinsmen, holding little in lordships, 
but following one head of their race (caput 
progenei— ceann cinnidh) as chief, with their 
friends and dependents." The lordships were 
held, alas ! by foreigners to them in race and 

The Clan Chattan were the native Celtic in- 
habitants of Badenoch. There are traditional 
indications that they came from the west— from 
Lochaber, where the MS. histories place the old 
Clan Chattan lands. The same authorities rs- 
oord that, for instance, the Macbeans came from 
Lochaber in the 14th century, " after slaying 
the Eed Comyn's captain of Inverlochy," and 
put themselves under the protection of Mackin- 
tosh; and this is supported by the tradition still 
preserved among the Eothiemurchus Macbeans, 
whose ancestor, Bean Cameron, had to fly 
Lochaber owing to a quarrel and slaughter 
arising from the exaction of the " bo ursainn," 
or probate duty of the time. It may be too bold 
to connect this eastern movement of Clan 
Chattan with the advancing tide of Scotic con- 
quest in the 8th century, whereby the Pictish 


kingdoms aiid the Pictish language were over- 
thrown. That thePicts inhabited Badenoch is 
undoubted : the place names amply prove that, 
for we meet with such test prefixes as Pet 
(PitowTie, Pictchirn, Pitmean) and Aber (Aber- 
arder), and other difficulties of topography un- 
explainable by the Gaelic language. As in most 
of Scotland, we have doubtless to deal, first, 
with a pre-Celtic race or races, possibly leaving 
remnants of its tongue in such a river name as 
Feshie, then the Pictish or Caledonian race of 
Celtic extraction, and, lastly, the Gaelic race 
who imposed their language and rule upon the 
previous peoples. The clan traditions are sup- 
ported in the matter of a western origin for the 
Clan Chattan by the genealogies given in the 
Edinburgh 1467 MS., which deduces the chief 
line from Ferchar Fota, King of Dalriada, in the 
7th century. 

The name Cattan, like everything connected 
with the early history of this clan, is obscure, 
and has, in like manner, given rise to many 
absurd stories and theories. As a matter of 
course, the classical geography of Europe has 
been ransacked, and there, in Germany, was a 
people called Chatti, which was taken as pro- 
nounced Catti ; but the ch stands for a sound like 
that in loch. The name appears as Hesse for 
Hatti. It was never Katti, be it remembered. 
Yet the Catti are brought from Germany to 


Sutheiiandshire, which in GaeHc is Cataibh, 
older Cataib — a name supposed thus to be de- 
rived from the Catti. Cataibh is merely the 
dative plural of cat (a cat), just as Gallaibh 
(Caithness) is the same case of Gall (a stranger, 
Norseman). The Cat men dwelt in Sutherland- 
shire; why they were called the Cats is not 
known. Clan Chattan is often said to be 
originally from Sutherland, but, beyond the 
similarity of name, there is no shadow of evid- 
ence for the assertion. Others again, like Mr 
Elton, see in the name Catan, which means, un- 
doubtedly, " little cat," rehcs of totemism; this 
means neither more nor less than that the pre- 
Christian Clan Chattan worshipped the cat, from 
whom, as divine ancestor, they deemed them- 
selves descended. We might similarly argue 
that the Mathesons — Mac .Mhath-ghamhuin or 
Son of the Bear — were a " bear " tribe, a fact 
which shows how unstable is the foundation on 
which this theory is built. In fact, animal 
names for men were quite common in early 
times. The favourite theory — and one coun- 
tenanced by the genealogies — connects the Clan 
Chattan, like so many other clans, with a 
church-derived name. The ancestor from 
whom they are represented as deriving their 
name is Gilhcattan Mor, who lived in the lltli 
century. His name signifies Servant of Catan, 



that is, of St Catan ; for people were named after 
saints, not directly, but by means of the prefixes 
Gille and Maol. At least, that was the early 
and more reverent practice. That there was a 
St Catan is evidenced by such place names as 
Kilchattan (in Bute and Lung), with dedication 
of churches at Gigha and Colonsay. His date 
is given as 710, but really nothing is known of 
him. This is probably the best explanation of 
the name, though the possibility of the clan 
being named after some powerful chief called 
Catan must not be overlooked. The crest of the 
cat is late, and merely a piece of mild heraldic 

It is only about or after 1400 that we come 
on anything like firm historical ground in the 
genealogy and story of our chief Highland clans. 
This is true of the Grants and the Camerons, and 
especially true of the Clan Chattan. Every- 
thing before that is uncertainty and fable. The 
earliest mention of Clan Chattan — and it is not 
contemporary but fifty years later — is in con- 
nection with the fight at the North Inch of Perth 
in 1396, and here historians are all at sixes and 
sevens as to who the contending parties really 
were. The battle of Invernahavon (1386?) and 
the fight at Clachnaharry (1454) are mere tra- 
ditions, and the battle in 1429 between Clan 
Chattan and Clan Chameron, in which the 
former nearly annihilated the latter, is recorded 


by a writer nearly a century later (1521). In 
fact, the first certain contemporary date is that 
of Mackintosh's charter in 1466 from the Lord 
of the Isles, where he is designated Duncan Mac- 
kintosh, " capitanus de Clan Chattan," and next 
year as " chief and captain " of Clan Chattan, 
in a bond with Lord Forbes. Henceforward, 
Clan Chattan is a common name in public his- 
tory and private documents. It comprised in 
the period of its comparative unity (circ. 1400- 
1600) some sixteen tribes or septs : these were 
the Mackintoshes, Macphersons, Davidsons, 
Cattanachs, Macbeans, Macphaiis, Shaws, 
Farquharsons, Macgillivrays, Macleans of Doch- 
garroch, Smiths, Macqueens, Gillanders, Clarks, 
&c. Of this confederation, Mackintosh was for, 
at least, two centuries " captain and chief/' ds 
all documents, public and private, testify. 
These two centuries (circ. 1400 to 1600) form 
the only period in which we see, under the light 
of history, the Highland clans in their full 

The 17th century made sad havoc in the 
unity of Clan Chattan. Huntly, ever an eneiiiy 
<o Mackintosh, "banded" in 1591 the Mac- 
phersons to his own person, and, by freely 
granting charters to them, made them inde- 
pendent, and detached them from Mackintosh. 
Macpherson of Cluny claimed to be head of the 
Macphersons, and in 1673 styled himself 


" Duncan jVPpherson of Cluney for himself, and 
taking burden upon him for the heall name of 
M^phersons and some others called old Clan- 
chattan as cheeffe and principall man thereoff, '' 
in a bond with Lord Macdonell of Morar. In 
support of this claim, the Macphersons appealed 
to the old genealogies, which represented Mac- 
kintosh as getting the Clan Chattan lands by 
marriage with the heiress in 1291, and which 
further showed that Cluny was the heir male 
descendant of the old Clan Chattan chiefs. Tiie 
case in its solemn absurdity of appeal to 
genealogies reminds one of a like appeal placed 
before the Pope in the claims of King Edward 
upon the throne of Scotland. He claimed the 
Scottish crown as the direct successor of Brutus 
and Albanactus, who lived in Trojan times, 
every link of genealogy being given, while the 
Scots repelled this by declaring that they were 
descended from Gathelus husband of Scota, 
daughter of the Mosaic King of Egypt ; and here, 
too, all the genealogical links could have been 
given. Neither doubted the genuineness of 
each others' genealogies ! So with the Mackin- 
tosh-Macpherson controversy about the chief- 
ship of Clan Chattan. They each accept each 
others' genealogies without suspicion or demur. 
And yet the manufacture of these and like 
genealogies was an accomplished art with Gaelic 
seanachies whether Irish or Scottish. We even 


see it going on under our very eyes. The early 
chiefs of Lochiel are the de Camhruns of the 
13th and 14th century records — hsts and other 
documents — impressed into the Cameron 
genealogy, which is doubtless correctly given m 
the 1467 MS. Again, the Macpherson 

genealogy in the Douglas Baronage is in several 
cases drawn from charters granted to wholly 
different famihes. Dormund Macpherson, 12th 
chief, gets a charter under the great seal from 
James IV. ; but the charter turns out to be one 
granted to a Dormund M'Pherson in the Lord- 
ship of Menteith, not of Badenoch ! John, 14th 
of Cluny, who " was with the Earl of Huntly at 
the battle of Glenlivet," as the veracious 
chronicler says, to add a touch of realism to his 
bald genealogical account, gets a charter of the 
lands of Tullich, &c., lands which lie in Strath- 
nairn, and he turns out to be a scion of the well- 
known family of Macphersons of |Brin ! 
Similarly John, 15th of Cluny, is son of the 
foregoing John of Brin; and Ewen, 16th of 
Cluny, who gets a charter in 1623 of the lands 
of Tullich, &c., is a cousin of Brin. Donald, 
17th of Cluny, who gets a charter in 1643, turns 
out to be Donald Macpherson of Nuid. And all 
this time another and a correct genealogy of the 
Cluny family had been drawn up by Sir ^Eneas 
Macpherson towards the end of the 17th cen- 
tiry, which must surely have been known to the 


writer.^ During all the period of 14th to 16th 
chief here given, there was only one man in 
Cluny, and his name was Andrew Macpherson, 
son of Ewen. 

The name Mackintosh signifies the son of 
the tbiseach or chief, which is Latinised by 0' 
Flaherty as " capitaneus sen praecipuus dux." 
The Book of Deer makes the relationship of 
tbiseach to other dignitaries quite plain. There 
is first the King; under him are the mormaers 
or stewards of the great provinces of Scotland, 
such as Buchan, Marr, and Moray; and next 
comes the tbiseach or chief of the clan in a par- 
ticular district. The two clans in the Book of 
Deer are those of Canan and Morgan, each with 
a tbiseach. This word is represented oftenest 
in English in old documents by thane, which, 
indeed, represents it with fair accuracy. 
Tbiseach is the true Gaelic word for " chief," 
but it is now obsolete, and there is now no true 
equivalent of the word ' ' chief ' ' in the 
language at all. And here it may be pointed 
out that the word chief itself was not at once 
adopted or adapted for this particular meaning 
of chief of a Highland clan. As we saw, the 
word at first employed was " captain," then 
" captain and chief," " captain, chief, and 
principal man," " chief and principal," &c., 

1 See Mr Fraser-Mackint-osh'.s Dunachton, pp. 46-49, for a 
full exposS of this remarkable piece of manufacture. 


the idea finally settling down as fully represented 
by the word "chief" in the 16th century. 
Skene's attempt to argue that captain denoted a 
leader temporarily adopted, leading the clan for 
another, or usurping the power of another, 
while chief denoted a hereditary office, is con- 
demned by his own evidence, and by the weight 
of facts. Besides, words do not suddenly spring 
into technical meanings, nor could chief acquire 
the definite meaning applicable to Highland 
chief ship, but by length of time and usage for 
this purpose. Hence arose the uncertainty of 
the early terms applied to the novel idea pre- 
sented by Highland clans. The word clan itself 
appears first in literature in connection with 
Clan Chattan, or rather Clan Qwhewyl, at the 
North Inch of Perth, where Wyntown speaks of 
'■ Clannys two." The Gaelic word clan had to 
be borrowed for want of a native English term; 
why should we then wonder at the idea of 
toiseach being rendered first by captain, and 
latterly by chief? 

The Mackintosh genealogies, dating from the 
1 7th century, represent the family as descended 
fiom Macduff, thane of Fife, as they and Fordun 
call him. Shaw Macduff, the second son of 
Duncan, fifth Earl of Fife, who died in 1154, in 
an expedition against the people of Moray in 
1160, distinguished himself, and received from 
the King lands in Petty, and the custody of 


Inverness Castle. Here he was locally known 
as Shaw Mac an Toiseich, " Shaw, the son of 
the Thane." He died in 1179, and was suc- 
ceeded by (2) Shaw, whose son was (3) 
Ferchard, whose nephew was (4) Shaw, whose 
son was (5) Ferchard, whose son was (6) Angus, 
who in 1291 married Eva, heiress of Clan 
Chattan, and thus got the Clan's land? in 
Lochaber. So far the genealogy. It is a pretty 
story, but it sadly lacks one thing — verisimili- 
tude. Macduff was not toiseach of Fife. In 
the Book of Deer he is called comes, the then 
Gaelic of which was mormaer, now moirear. 
Shaw Macduff would infallibly, as son of the 
Earl of Fife, have been called Mac Mhoireir. 
With those who support this Macduff genealogy, 
no argument need be held; like the humorist of 
a past generation, one would, however, like to 
examine their bumps. The statement that the 
Mackintoshes were hereditary constables of 
Inverness Castle is totally baseless and false. 
At the dates indicated (12th century) we believe 
that the Mackintoshes had not penetrated so far 
north as Petty or Inverness, and that we should 
look to Badenoch as their place of origin, and 
their abode at this time. Unfortunately docu- 
ments in regard to the early history of Badenoch 
are rare, but an entry or two in the Eegistrum 
of Moray Diocese may help us. In 1234, 
Walter Comyn, Earl of Monteith, comes to an 


agreement with the Bishop of Moray, in regard 
to Kincardine, and Fercard, son of Seth, is a 
witness, and in the very next document, also 
one of Walter Comyn's, of the same date, ap- 
pears a witness called Fercard " Senescalli de 
Badenoch," that is " steward of Badenoch." 
We are quite justified in regarding him as the 
person mentioned in the previous document as 
Fercard, son of Seth. Now^ one translation of 
toiseach is steward or seneschal — the person in 
power next the mormaer or earl. We may, 
therefore, conclude that this Ferchard was 
known in Gaelic as Ferchard Toiseach. Similarly 
in 1440 we meet with Malcolm Mackintosh, chief 
of the clan, as " ballivus de Badenoch," a title 
of equal import as that of seneschal. We 
should then say that the Mackintoshes derived 
their name from being toiseachs of Badenoch, the 
head of the old Celtic clan being now under the 
new non-Celtic mormaer or earl Walter Comyn. 
The ease with which the name Mackintosh might 
arise in any place where a clan and its tbueacl' 
existed explains hov/ we meet with Ma(?kiji- 
toshes, for instance, in Perthshire, who do not 
belong to the Clan Chattan. Thus there were 
Mackintoshes of Glentilt, which was held as an 
old thanage, and whose history as such is well 
known. Similarly we may infer that the Mac- 
kintoshes of Monivaird were descendants of the 
old local Toiseachs or Thanes. The Mackintosh 


genealogists have of course annexed them to the 
Clan Chattan stock with the utmost ease and 
success. In 1456, John of the Isles granted to 
Somerled, his armour bearer, a davoch of the 
lands of Glennevis, with toiseachdorship of most 
of his other lands there, and in 1552 this grant 
is renewed by Huntly to " dilecto nostro 
Donaldo MacAhster M'Toschd," that is, 
Donald, son of Alister, son of Somerled, the 
toiseach or bailif, named in 1456. This shows 
how easily the name could have arisen. 

Skene, while unceremoniously brushing 
aside the Macduff genealogy, advances 
hypothetically a different account of the origin 
of the Mackintoshes. In 1382, the Lord of 
Badenoch is asked to restrain Farchard Mac- 
Toschy and his adherents from disturbing the 
Bishop of Aberdeen and his tenants in the land 
of Brass or Birse, and to oblige him to prosecute 
his claim by form of law. Skene thinks that 
Farchard, whom he finds in the 1467 MS. as one 
of the "old" Mackintoshes, was descended 
from the old thanes of Brass, and that hence 
arose his name and his claim. Being a vassal 
of the Wolf's, he was a Badenoch man too. 
Rothiemurchus was a thanage, and the con- 
nection of the Mackintoshes with it was always 
close. Alexander Keir Mackintosh obtained 
the feudal rights to Rothiemurchus in 1464, and 
a few years later he styles himself " Thane of 


Rothiemurchus." Skene then suggests that 
Birse and Rothiemurchus might have anciently 
been in the hands of the same toiseach or thane, 
and that from him the Mackintoshes got their 
name. We have suggested that the name arose 
with Ferchard, son of Seth or Shaw, who was 
toiseach under the Earl Walter Comyn in 1234, 
and his name appears in the 1467 MS. genealogy 
as well as in the Mackintosh genealogies. 

That a revolution took place in the affairs of 
Clan Chattan, with the overthrow or extrusion 
of the direct line of chiefs, in the half century 
that extends from about 1386 to 1436, is clear 
from two sources — first, from the 1467 MS., 
and, second, from the Mackintosh history. The 
latter acknowledges that Ferquhard, 9th chief, 
was deposed from his position, which was given 
to his uncle Malcolm. The reason why he had 
to retire was, it is said, the clan's dissatisfaction 
with his way of managing affairs ; but the matter 
is glossed over in the history in a most unsatis- 
factory manner. If this was the Ferchard men- 
tioned in 1382 as giving trouble to the Bishop 
of Aberdeen, it is most unlikely that he was an 
incapable man; in fact, he must have been quite 
the opposite. He is doubtless the same person, 
for he is given also in the 1467 MS. genealogy. 
But further confusion exists in the Mackintosh 
account. Malcolm, 10th Mackintosh, who dies 
in 1457, is grandson through Wilham 7th (died 


1368) of Angus who married Eva in 1291, the 
three generations thus lasting as chiefs from 
1274 to 1457, some 183 years! Malcohii was 
the son of William's old age, and his brother, 
Lachlan 8th, was too old to take part in the 
North Inch fight in 1396, sixty years before his 
younger brother died ! This beats the Fraser 
genealogy brought forward lately by a claimani 
to the Lovat estates. It is thus clear that there 
is something wrong in the Mackintosh genealogy 
here, corresponding doubtless to some revolu- 
tion in the clan's history. And this is made 
clear when we consult the Edinburgh Gaelic MS. 
of 1467, which gives the genealogies of High- 
land clans down till about 1450. Here we 
actually have two genealogies given, which 
shows that the chiefship of the Mackintoshes or 
Clan Gillicattan was then either in dispute or a 
matter of division between two families. We 
print the two 1467 hsts with the Mackintosh MS. 
genealogy between them, in parallel columns, 
supplying dates where possible : — 

1467 MS. Mackintosh History. I46? MS. 

William and Donald (12) Fercliar (d. 1514) Lochlan 

William (9) Ferchar (11) Duncan (d. 1496) Suibne 

Fercliar (1382) (8) LachUn & (10; Malcolm (d. 1457) Shaw 

Wiiliam (7) W^illiam id. 1368) Leod 

Gii:amichol (6) An-us (d. 1345) Scayth (1338) 

Ferchar (1 '234) (5) Ferchar (d. 1274) Ferchard 

Shaw (4) Shaw (d. 1265) Cilchrist 

Gilclirist William Malcolm 

Aigcol (2) Shaw (d. 1210) DoualdCamgiila 

Ewcn (1) Shaw (d. 1179) Mureach 

Macduff (d. 1154) Suibiio 

Earl of Fife ^ Tead (Shaw) 

Keill " Nachtain 

[Gillicattan ?] Qillicattau 


The similarity between the 1467 first list and 
that of the Mackintosh history is too striking to 
be accidental, and we may take it that they 
purport to give the same genealogy. There are 
only two discrepancies from about 1400 to 1200 
between them. Ferchar 9th is given as soi; of 
Lachlan in the Mackintosh history, whereas the 
1467 Hst makes him son of WilHam, not grand- 
son. The 6th Mackintosh in the one hst is 
Gillamichael, and in the other he is called 
Angus. Perhaps he had borne both names, for 
Gillamichael means " servant of St Michael," 
and might possibly be an epithet. Mr Fraser- 
Mackintosh has drawn the writer's attention to 
a list of names published in Palgrave's " Docu- 
ments and Eecords " of Scottish History (1837); 
this is a list of some ninety notables who, aboal 
1297, made homage or submission to Edward I., 
and among them is Anegosius Maccarawer, or 
Angus Mac Ferchar, whom Mr Fraser-Mackin- 
tosh claims as the 6th of Mackintosh. There 
are only two other " Macs " in the list, and 
Maccarawer is, no doubt, a Highlander, and 
possibly a chief, and, perhaps, the chief of Mac- 
kintosh.^ In any case, in the middle of the 
15th century, the direct line of Mackintoshes 
was represented by William and Donald, sons of 
William, whereas the chief de facto at the time 

1 Angus McErchar here is chief of the Lamonts in Argyle. 


was undoubtedly Malcolm Mackintosh. How 
he got this position is a question. 

The second Hst in the 1467 MS. is a puzzle. 
Mr Skene called it the genealogy of the " old ' ' 
Clan Chattan : Why, is not clear. Scayth, son 
of Ferchard, is mentioned in 1338 as the late 
Scayth who possessed a " manerium " at the 
" stychan " of Dalnavert. Mr Skene thinks 
that he was of the Shaws of Rothiemurchus, and 
that this is their genealogy; and this may be 
true, but what comes of his earlier theories in 
regard to the Macphersons as being the ' ' old 
family here represented? Theories held in 
1837 were abandoned in 1880; but in this Mr 
Skene could hardly help himself, considering the 
amount of information that has since appeared 
in the volumes of such Societies as the " Spald- 
ing Club," bearing on the history of the 
Moravian clans, and especially on that of Claa 

The turmoil in the Clan Chattan, which 
changed the chief ship to another line, must be 
connected more especially with the events which 
took place when King James came North, in 
1427, when part of the clan stood by the King 
and part by the Lord of the Isles. We find in a 
document preserved in the Kilravock papers, 
that King James grants a pardon to certain of 
the Clan Chattan, provided they really do attach 
themselves to the party of Angus and Malcolm 


Mackintosh; and this shews that Malcolm, who 
was afterwards chief, stood by the king, and 
received his favours. Angus possibly was his 
brother, for a depredating rascal of the name 
of Donald Angusson, supported by Lachlan 
" Badenoch," son of Malcolm, evidently Lach- 
lan' s cousin, gives trouble to various people to- 
wards the end of the century. In any case, 
Malcolm Mackintosh emerged from the troubles 
that were rending the clan victorious, and his 
son Duncan was as powerful a chief as lived m 
the North in his day. 

How much the Clan Battle at Perth, in 1396, 
had to do with the changes in the Clan Chattan 
leadership it is hard to say. It is accepted as 
certain that the Clan Chattan had a hand in the 
fight, for the later historians say so, and the con- 
temporary writer Wyntoun mentions the chiefs 
on both sides, and one of these bears the name 
of Scha Ferchar's son, which is an unmistake- 
ably Mackintosh name. He says, in Laing's 
edition : — 

" Tha thre score were clannys twa, 
Clahynnhe Qwhewyl, and Clachinya; 
Of thir twa Kynnys ware the men, 
Thretty agane thretty then. 
And thare thai had thair chifftanys two, 
Schir Ferqwharis sone wes ane of tha, 
The tothir Cristy Johnesone." 


The two clans here pitted against one another 
are the clans Quhele or Chewil, and Clan Ha or 
Hay, or, according to some, Kay. Boece has 
Clan Quhete, which Buchanan and Leslie im- 
prove into Clan Chattan. 

As so much theorising has taken place upon 
this subject already, and so many positive 
assertions have been made, it may at present 
serve the interests of historic science if we can 
reallv decide what clan names the above cannot 
stand for. First, there is Clan Quhele or 
Chewil. This clan is mentioned in 1390 as Clan 
Qwhevil, who, with the Athole tribes, made a 
raid into Angus, and killed the Sheriff. They 
are mentioned again in an Act of Parliament in 
1594 as among the broken clans, in the follow- 
ing sequence — Clandonochie, Clanchattane, 
Clanchewill, Clanchamron, &c. What clan 
they really were is yet a matter of dispute. The 
form Che will points to a nominative, Cumhal or 
Cubhal, or Keval, but no such name can be 
recognised in the Clan Chattan district, or neat 
it. Dughall or Dugald has been suggested, and 
the family of Camerons of Strone held as the 
clan referred to. But this, like so much in the 
discussion of this subject, forgets some very 
simple rules of Gaelic phonetics, which are not 
forgotten in the spoken language, and in the 
English forms borrowed from it. Ferninine 
names ending in n never aspirate an initial d of 


iJie next ivord. We have Clan Donnachie, Clan 
Donald, Clan Dugald, and so on, but never Clan 
Yonnachie or Yonald, or such. Similarly, Ckii 
Hay or Ha cannot stand for Clan Dai or David- 
sons. Let these simple rules of Gaelic phonetics 
be understood once for all, and we have made 
much progress towards a solution of the diffi- 
culty. The word Qwhevil evidently commences 
with a C. Skene suggests it is for Caimgilla, 
"one-eyed one," the epithet of Donald, 
Mureach's son, in the 1467 pedigree. But the 
m of cam is never aspirated. Again, as to Ha 
or Hay. The Clan Cameron are called, in the 
1467 MS. and other places, the " Clann 
Maelanfhaidh," the clan of the " servant 
of Storn;," a name preserved in the 
Macgillony of Strone, which originally was Mac 
Gille-anfhaidh, equivalent to Mael-anfhaidh in 

The name, however, that best suits the Eng- 
lish form is that of Shaw or Seadh, that is, Seth. 
There is really a difficulty about Mael-anfhaidh 
and his clan. The form ought to be either 
Clann-anfhaidh, which Wyntoun would give as 
Clahinanha or Clahan-anna, or it would be Clann 
Mhael-anfhaidh, a form which could not be mis- 
taken, were it handed down. The most popular 
theory at present is that the combatants were the 
Camerons and Mackintoshes, who were enemies 



for three centuries thereafter; the Mackintoshes 
were represented by the name of Clan Chewill, 
the chief being Shaw, son of Ferchar, of the 
Eothiemurchus branchy while the Camerons were 
the Clan Hay, with Gilchrist Mac Iain as chief. 
This is practically Skene's view, and it is the 
position taken up by Mr A. M. Shaw, the his- 
torian of the Mackintoshes. But the phonetics 
point to a struggle in which the Shaws were the 
chief combatants, the other side being Clan 
Kevil, and, on weighing all sides of the question, 
we are as much inclined to believe that it was the 
beginning of that struggle in the clan, which is 
represented by two lines of pedigree, and which 
latterly gave the chiefship even to a junior 
branch of one of the lines. 

How does the claim of the Cluny Macphersons 
for the chiefship of Clan Chattan stand in rela- 
tion to these historic facts? They do not appear 
at all in the historical documents, but tradition 
in the seventeenth and eighteenth centuries had 
enough to tell of their share in the crisis. At 
the battle of Invernahaven, fought against the 
Camerons, the Macphersons of Cluny claimed 
the right under Mackintosh as chief, but he un- 
fortunately gave this post of honour to the Clan 
Dai or Davidsons of Invernahaven ; and the Mac- 
phersons retired in high dudgeon. The battle 
was at first lost to Clan Chattan, but the Mac- 
phersons, despite anger, came to the rescue, 


and the Camerons were defeated. Then ensued 
a struggle, lasting ten years, for superiority 
between the Macphersons (Clan Chattan) and 
the Davidsons, the scene of which, in 1396, was 
shifted to the North Inch of Perth. These, th^ 
Macpherson tradition says, were the two clans 
that fought the famous clan fight. The Mac- 
phersonf^ claim to be descended from Gilhcattan 
Mor, progenitor of the Clan Chattan, by direct 
male descent, and every link is given back to 
the eleventh century, thus (omitting ' ' father 
of") — Gillicattan, Diarmid, Gilhcattan, Muirich, 
parson of Kingussie, whence they are ? ailed 
Clann Mhuirich, father of Gillicattan a'nd Ewen 
Ban, the former of whom had a son, Dougid 
Dall, whose daughter Eva, " the heiress of Clan 
Chattan," married Angus Mackintosh in 1291, 
and thus made him ' ' captain ' ' of Clan Chattan ; 
Ewen Ban was the direct male representative, 
then Kenneth, Duncan, Donald Mor, Donald Og, 
Ewen; then Andrew of Cluny in 1609, a real 
historic personage without a doubt. In this list, 
not a single name previous to that of Andrew 
can be proved to have existed from any docu- 
ments outside the Macpherson genealogies, ex- 
cepting only Andrew's father, Ewen, who is 
mentioned in the Clanranald Eed Book as grand- 
father of the heroic Ewen, who joined Montrose 
with three hundred of Clans Mhuirich and 
Chattan. The direct Gillicattan genealogy is 


given in the 1467 MS., and, such as it is, it has 
no semblance to the Macpherson hst. The fact 
is that the Macpherson hst previous to Ewan, 
father of Andrew, is purely traditional and 
utterly unrehable. The honest historian of 
Moray, Lachlan Shaw, says — " I cannot pre- 
tend to give the names of the representatives be- 
fore the last century. I know that in 1660 
Andrew was laird of Clunie, whose son, Ewan, 
was father of Duncan, who died in 1722 without 
male issue." By means of the Spalding Pub- 
lications, the Synod of Moray Records, and 
other documents, we can now supplement and 
add to Lachlan Shaw's information, though not 
much. Macpherson of Cluny is first mentioned 
in 1591 when Clan Farson gave their " band " 
or bond to Huntly. He is then called " Andrew 
Makfersone in Cluny," not of Cluny, be it ob- 
served, for he was merely tenant of Cluny at that 
time. This is amply proved by the Badenoch 
rental of 1603, where we have the entry — 
" Clovnye, three pleuches . . . Andro 
McFarlen {read Farsen)^ tenant to the haill." 
In 1609, Andrew had obtained a herit- 
able right to Cluny, for then he is 
called Andrew Macpherson of Cluny in 
the bond of union amongst the Clan 
Chattan, ' ' in which they are and is astricted 
to serve Mackintosh as their Captain and Chief." 

1 The MS. has farsen; " Farlen " is a misprint. 


Huntly had for long been trying to detach the 
Clan from Mackintosh by " bands," as in 1591 
and in 1543, and by raising the tenants to a 
position of independence under charter rights, 
which were liberally granted in the seventeenth 
century, and which proved fatal to the unity of 
Clan Chattan. But it was a wise policy, 
nationally considered, for in 1663-5, when Mac- 
kintosh tried to raise his Clan against Lochiel, 
some flatly refused asking cui bono; others 
promised to go if Mackintosh would help them 
to a slice of their neighbour's land, and Mac- 
pherson of Cluny proposed three conditions on 
which he would go — (1) if the Chiefs of the 
Macphersons hold the next place in the Clan to 
Mackintosh; (2) lands now possessed by Mac- 
kintoshes and once possessed by Macphersons 
to be restored to the latter; and (3) the assist- 
ance now given was not of the nature of a ser- 
vice which Mackintosh had a right to demand, 
but simply a piece of goodwill. When Mackin- 
tosh was in 1688 proceeding to fight the " last 
Clan battle ' ' at Mulroy against Keppoch, we are 
told that the " Macphersons in Badenoch, after- 
two citations, disobeyed most contemptuously." 
Duncan Macpherson, the Cluny of that time, 
had decided to claim chiefship for himself, and 
in 1672 he applied for and obtained from the 
Lord Lyon's Office the matriculation of his arms 
as Laird of Cluny Macpherson, and only true 


representative of the ancient and honourable 
family of Clan Chattan. Mackintosh, on hear- 
ing of it, objected, and got the Lord Lyon to 
give Macpherson " a coat of arms as cadets of 
' Clan Chattan.' " The Privy Council in the 
same year called him " Lord of Cluny and Chief 
of the Macphersons," but Mackintosh got them 
to correct even this to Cluny being responsible 
only for " those of his name of Macpherson 
descendit of his family," without prejudice 
always to the Laird of Mackintosh. In 1724 
Mackintosh and Macpherson came to an agree- 
ment that Mackintosh, in virtue of marrying the 
heiress of Clan Chattan in 1291, was Chief of 
Clan Chattan, Macpherson renouncing all claim, 
but there was a big bribe held out to him — he 
received the Loch Laggan estates from Mackin- 
tosh. In this way the egging on of Huntly, the 
reputation gained by the Macphersons in the 
Montrose wars and otherwise, and an absurd 
piece of pedigree, all combined to deprive Mac- 
kintosh of his rightful honour of Chief, and also 
of a good slice of his estate ! The renown 
gained by the Clan Macpherson in the Jacobite 
wars, compared to the supineness of the Mac- 
kintosh Chiefs, gained them public sympathy in 
their claims, and brought a clan, altogether un- 
known or ignored until the battle of Glenlivet 
in 1694, to the very front rank of Highland 
Clans in the eighteenth century. We see the 


rise of a clan and its ciiiefs actually take place 
in less than a century and a half, and that, too, 
by the pluck and bravery displayed by its chiefs 
and its members. 


The Ordnance Survey maps, made to the 
scale of six inches to the mile, contain for 
Badenoch some fourteen hundred names; but 
these do not form more than a tithe of the names 
actually in use or once used when the glens were 
filled with people, and the summer shealings 
received their annual visitants. Every knoll 
and rill had its name; the bit of moor, the bog 
or hlar^ the clump of wood {hadav), the rock or 
crag, the tiny loch or river pool, not to speak of 
cultivated land parcelled into fields, each and 
all, hovv'ever insignificant, had a name among 
those that dwelt near them. Nor were the 
minute features of the mountain ranges and far- 
away valleys much less known and named. The 
shealing system contributed much • to this last 
fact. But now many of these names are lost, 
we may say most of them are lost, with the loss 
of the population, and with the abandonment of 
the old system of crofting and of summer migra- 
tion [O the hills. The names given to those 
minute features of the landscape were and are 
comparatively easy on the score of derivation, 


though sometimes difficult to explain historically. 
For instance, Lub Mhairi, or Mary's Loop, is ihe 
name of a small meadow at Coilintuie, but vhr^ 
was the Mary from whom it got its name ? 

Of the fourteen hundred words on the 
Ordnance Maps, we may at once dismiss three- 
fourths as self-explanatory. Anyone with a 
knowledge of Gaelic can explain them; or any- 
one not so endowed but possessed of a Gaelic 
dictionary can by the use of it satisfactorily un- 
ravel the mystery of the names. Of the remain- 
ing fourth, most are easy enough as regards 
derivation, but some explanation of an historical 
character is desirable, though often impossible 
of being got. One of the most interesting names 
under this last category is that of Craig Righ 
Harailt, or the Crag of King Harold, which 
stands among the hills behind Dunachton; yet 
there is absolutely nothing known about this 
Scandinavian chief; even tradition halts m the 
matter. There are only some six score names 
where any difficulty, however slight, of deriva- 
tion can occur, and it is to these names that this 
paper will mostly devote itself. The oldest 
written or printed form of the name will be 
given, for often the difficulty of deriving a place- 
name yields when the oldest forms of it are 
found. We have fortunately some valuable 
documents, easily attainable, which throw light 
on some obscure names. Among these are the 


Huntly Rental for the Lordship of Badenoch for 
1603/ and Sh^ R. Gordon of Straloch's map of 
Braidalbane and Moray, which was pubhshed in 
Blaeu's Atlas in 1662, and which contains a full 
and intelligent representation of Badenoch. 
The Badenoch part of this map is reproduced 
along with this paper for the sake of illustrating 
it. It was made about the year 1640. 

First, we shall deal with the name of the dis- 
trict and the names of the principal divisions of 
it, and thereafter consider the nomenclature of 
the leading features of the country, whether 
river, loch, or mountain, following this with a 
glance at the names of farms and townships, and 
at the other points of the landscape that may 
seem to require explanation. The name of the 
district first claims our attention. 

Badenoch. — In 1229 or thereabouts the 
name appears as Badenoch in the Registrum of 
Moray Diocese, and this is its usual form there ; 
in 1289, Badenagh, Badenoughe, and, in King 
Edward's Journal, Badnasshe; in 1366 we have 
Baydenach, which is the first indication of the 
length of the vowel in Bad-; a 14th century map 
gives Baunagd; in 1467, Badyenach; in 1539, 
Baidyenoch; in 1603 (Huntly Rental), 
Badzenoche; and now in Gaelic it is Baideanach. 
The favourite derivation, first given by Lachlan 

1 Spalding Club Miscellany, vol. iv. 


Shaw, the historian of Moray (1775), refers it to 
badan, a bush or thicket; and the Muses have 
sanctioned it in Calum Dubh's expressive hne in 
his poem on the Loss of Gaick (1800) — 

" 'S bidb. muiru auu an Dutliaich uam Badau " 
(And joy shall be in the Land of Wood-clumps). 

But there are two fatal objections to this deriva- 
tion; the a of Badenoch is long, and that of 
hadan is short; the d of Badenoch is vowei- 
fianked by " small " vowels, while that of hadan 
is flanked by " broad " vowels and is hard, the 
one being pronounced approximately for English 
as hah-janach, and the other as baddanacli. 
The root that suggests itself as contained in the 
word is that of hath or badh (drown, submerge), 
which, with an adjectival termination in de, 
would give haide, " submerged, marshy," and 
this mi^'ht pass into haidean and haideanach, 
" marsh or lake land." That this meaning suits 
the long, central meadow land of Badenoci', 
which once could have been nothing else than a 
long morass, is evident. There are several 
places in Ireland containing the root badh 
(drown), as Joyce points out. For instance, 
Bauttagh, west of Loughrea in Galway, a marshy 
place; Mullanbattog, near Monaghan, hill sum- 
mit of the morass; the river Bauteoge, in 
Queen's County, flowing through swampy 
ground; and Currawatia, in Galway, means the 


inundated curragh or morass. The neighbour- 
ing district of Lochaber is caUed by Adamnan 
Stagnum Aporicum, and the latter term is hkely 
the Irish ah ar (a marsh), rather than the Pictisli 
a5er (a coniluence); so that both districts may be 
looked upon as named from their marshes. The 
divisions of Badenoch are three — the parishes of 
Alvie, Kingussie and Insh, and Laggan. 

Alvie. — Shaw says it is a " parsonage dedi- 
cated to St Drostan." Otherwise w^e should 
have at once suggested the 6th century Irish 
saint and bishop called Ailbe or later Ailbhe, 
whose name suits so admirably, that, even de- 
spite the Drostan connection, one would feel in- 
clined to think that the parish is named after St 
Ailbhe. In the middle of the 14th century the 
parish is called Alveth or Alwetht and Alway, 
amd Alvecht about 1400, in 1603 Alvey and 
Aluay, and in 1622 AUoway. The name, with 
the old spelling Alveth, appears in the parish of 
Alvah in Banffshire, and no doubt also in that of 
Alva, another parish in Stirlingshire. Shaw 
and others connect the name with ail (a rock), 
but do not explain the c or bh in the name. 
Some look at Loch Alvie as giving the name to 
the parish, and explain its name as connected 
with the flower ealhhaidh or St John's wort, a 
plant which it is asserted grows or grew around 
its bank. The learned minister of Alvie in Dis- 
ruption times, Mr Macdonald, referred the name 


of the loch to Eala-i oi' Swan-isle Loch, but un- 
fortunately there is no Gaelic word i for an 
island, nor do the phonetics suit in regard to the 
bli or V. The old Fenian name of Alnihu or 
Almhuinn, now Allen, in Ireland, the seat of 
Fionn and his Feinn, suggests itself, but the 
termination in n is wanting in Aivie, and this 
makes the comparison of doubtful value. 

Insh. — Mentioned as Inche in the Moray 
Eegistrum in 1226 and similarly in 138U and in 
1603. The name is derived from the knoll on 
which the church is built, and which is an island 
or innis when the river is in flood. Loch Insh 
takes its name from this or the other real island 
near it. The parish is a vicarage dedicated to 
" St Ewan," says Shaw; but, as the name of 
the knoll on which the church stands is Tom 
Eunan, the Saint must have been Eonan or 
Adamnan, Columba's biographer, in the 7th 
century. The old bell is a curious and rare 
relic, and the legend attached to it is one of the 
prettiest told in the district. The bell w^^s 
stolen once upon a time, and taken to the south 
of the Grampians, but getting free, it returned 
of its own accord ringing out as it crossed the 
hills of Drumochter, ' ' Tom Eonan ! Tom 

Kingussie. — In Gaelic — Cinn-ghiuhhsaich — 

(at) the end of the fir-forest " ; cinn being the 

locative of ceann (head) and giuhhsach being a 


" fir-forest." The oldest forms of the name are 
Kynguscy (1103-11?), Kingussy (1208-15), Kin- 
gusy (1226), Kingucy (1380), Kingusy (1538), 
and Kyngusie (1603). It is'a parsonage dedicated 
toStCohmiba (Shaw). According to Shaw, there 
was a Priory at Kingussie, founded by the Earl 
of Huntly about 1490. The Prior of Kingussie 
was the tenant of the Churcli croft in 1565 — 
Reg. of Moray, p. 450. 

Laggun.—'' A mensal church dedicated to 
St Kenneth " (Shaw). The name in full is 
Laggan-Choinnich, the lagan or " hollow of 
Kenneth." The present church is at Laggan 
Bridge, but the old church was at the nearest 
end of Loch Laggan, where the ruins are still 
to be seen. It is mentioned in 1239 as Logyn- 
kenny (E.M.), and Logykenny shortly before, 
as Logachnacheny and Logykeny in 1380, 
Logankenny in 1381 (all from R.M.), and 
Lagane in 1603 (H.R.). The Gaehc word 
" lagan " is the diminutive of " lag," a hollow. 

We now come to the leading natural features 
of the country, and deal first with the rivers and 
loche of Badenoch. A loch and its river gener- 
ally have the same name, and, as a rule, it is the 
river that gives name to the loch. A prominent 
characteristic of the river names of Badenoch, 
and also of Pictland, is the termination ie or y. 
We meet in Badenoch with Feshie, Trommie, 
Markie, and Mashie, and not far away are- 


Bennie, Druie, (reldie, Garry, Bogie, Gaudie, 
Lossie, Urie, and several more. The termina- 
tion would appear to be that given by Ptolemy 
in several river names such as Nov-ios, Tob-ios, 
Libn-iocS, &c., which is the adjectival termina- 
tion ios; but it has to be remarked that the 
modern pronunciation points to a termination in 
idh, Zeuss's primitive adi or idi; Tromie in 
Gaelic is to be spelt Tromaidh, and Feshie as 
Feisidh. We first deal with the so-called 
" rapidest river in Scotland." 

The Speif. — The Highlanders of old had a 
great idea of the size of the Spey, and also of the 
Dee and Tav. There is a Gaehc saying which 
runs thus : — 

Spe, De, agus Tatha, 
Tri uisgeachau 's mo fo'u athar. 

This appears in an equally terse English form — 

The three largest rivers that be 
Are the Tay, the Spey, and the Dee, 

" The river Spey is spoken of as ' she,' and has 
the character of being ' blood thirsty.' The 
common belief is that ' she ' must have at least 
one victim yearly." — W. Gregor, Folklore, III., 
72. In Norse literature the name appears as 
Spas (13th century); we have the form Spe in the 
"Chronicles" (1165); Spe (1228, &c.); Spee 
(Bruce' s Charter to Eandolph) and Spey (1451 
and 1603). But the Spey is regarded as repre- 


senting physically and etymologically Ptolemy's 
river Tvesis or Tvsesis. Dr Whitley Stokes says . 
— " Supposed to be Ptolemy's Tvesis; but it 
points to an original Celtic squeas, cognate with 
Ir. sceivi (vomo), W. chivy d (a vomit). For 
the connection of ideas, cf. Pliny's Vomanus, a 
river of Picenum. The river name Spean may 
be a diminutive of Spe." The changing of an 
original s git to S2:>, instead of the true Gaelic form 
sg or so, indicates that the name is Pictish. Tiie 
Spean is doubtless a diminutive arising from a 
form spesona or spesana. 

The Dulnan; in Gaelic Tuilnean, Blaeu's 
map Tiilnen. It falls into Spey near Broomhih 
Station. The root is tuil, flood; the idea being 
to denote its aptness to rapid floods. 

Feshie; Gaelic Feisidh. Its first appearance 
in charters is about 1230, and the name is 
printed Ceffy, evidently for Fessy. If it is 
Celtic, its earliest form was Vestia, from a root 
ved, which signifies "wet," and which is the 
origin of the English word wet and water. That 
Feshie is Celtic and Pictish may be regarded as 
probable when it is mentioned that in Brecon- 
shire there is a river Gwesyn, the root of the 
name being gives (for vest), meaning " what 
moves " or " goes." 

Tromie; Gaehc Trom{a)idh. In 1603 it is 
called Tromye. The Gaehc name for dwarf 
elder is tromayi, which appears in Irish as tro^n 


or tromm, with genitive truimm. It gives its 
name to Trim in Meatli, which in the 9th cen- 
tury was called Vadiim Truimm, or Ford of the 
Elder-tree. Several other Irish place-names 
come from it. In Badenoch and elsewhere in 
the Highlands, we often meet with rivers named 
after the woods on their banks. Notably is so 
the case with the alder tree, Fearna, which 
names numerous streams, and, indeed, is found 
in old Gaul, for Pliny mentions a river called 
Vernodubrum. Hence Tromie is the Elder-y 
Biver; while Truim, which is probably named 
after the glen, Glen-truim — " Glen of the 
Elder '"' — takes its name from the genitive of 
tromm. Compare the Irish Cala-truim, the 
liollow of the elder. Glen-tromie is the first 
part of the long gorge that latterly becomes 
Gaick, and, in curious contrast to the ill fame 
of the latter in poetry, it appears thus in a well- 
known verse : — 

Gleaan Tromaidh nan siautan 
Learn bvi mhiann bhi 'nad fha.sgatli. 

Par am faig-hinu a' bhroiglileag. 
An oiglueag 'b an d«arcag, 

Cnothan donn air a' challtuinn, 
'S iasg dearg air na h-easan. 

Guinag, Gitynack, Guinach, or Gynach (pro- 
nbunced in Gaelic Goi{hh)meag), falls into the 
Spey at Kingussie. It is a short, stormy stream- 
let. All sorts of derivations have been offered; 


the favourite is guanag, pretty, but, unfor- 
tunately, it does not suit the phonetics of 
Goi-neag. The name points to primitive forms 
Hke gohni- or gomni-, where the o may have 
been a, and the latter form, read as gamni-, 
would give us the root gam, which in old Gaelic 
means "winter." Hence the idea may be 
" wintry streamlet." Compare Coire Ghoibli- 
nidh in Rosskeen — W. J. Watson, Place Names 
of Ross and Cromarty. 

The C alder : in GaeHc Cal(l)adar. This 
river and lake name recurs about a dozen times 
in Pictland and the old Valentia province be- 
tween the Walls, and there is a Calder river in 
Lancashire. Cawdor and its Thanes probably 
give us the earliest form of the word, applied to 
the Nairnshire district. This is in 1295 Kaledor; 
in 1310, Caldor; and in 1468, Caudor. But the 
Gaelic forms persist in other places, as in Aber- 
Callador (1456) in Strathnairn. These forms 
point to an older Cal-ent-or, for ent and ant 
become in Gaelic ed or ad, earlier et or at. In 
the Irish Annals mention is made of a battle, 
fought, it is supposed, in the Carse of Falkirk, 
called the battle of Calitros, and certain lands 
near Falkirk were called in the 13th century 
Kalentyr, now Callendar. Not far away are 
several Calder waters. Compare also Callatei 
burn, Aberdeenshire, pronounced as spelt; in 



1564, Auchinquhillater, a farm, now i\.uchalter; 
Calder river in Renfrewshire. The root is evi- 
dently cal (sound, call), as in Latin Calendae, 
and English Calendar, borrowed, like the Gaelic 
equivalent word Caladair, from the Latin Calen- 

The Truim. See under the heading of 

The Mashie; Masie (1603), in Gaelic 
Mathaisidh, pronounced Mathisidh. Strath- 
mashie is famous as the residence of Lachlan 
Macpherson, the bard, the contemporary and 
coadjutor of James Macpherson of Ossianic re- 
nown. The bard's opinions of the river Mashie 
are still handed down; these differed according 
to circumstances. Thus he praised the river : — 

Mathaisidh gheal, bhoidheach gheal, 

Mathaisidh gheal, bhoidheach gheal, 

Bu chaomh learn bhi laimh riut. 

But after it carried away his corn he said : — 

Mathaisidh dhvibh, fhrogach dhubh, 
Mathaisidh dhubh, fhrogach dhubh. 
Is mor rinn thu chall orm. 

The derivation of the name is obscure. 
Mathaisidh could come from mathas, goodness, 
but the meaning is not satisfactory. We might 
think of maise, beaut}^ but it has the vowel short 
in modern Gaelic, though Welsh maws, 
pleasant, points to a long vowel or a possible 
contraction in the original. 


The Markie; Gaelic Marcaidh. Streams and 
glens bearing the name Mark and Markie occur 
in Perthshire, Forfarshire, and Banffshire. The 
first tributary of the Feshie is Alit Mharkie, at 
the mouth of which was of old Invermarkie, an 
estate held by the Campbells of Cawdor in the 
15th and 16th centuries. The root is doubtless 
marc, a horse. 

The Pattack; in Gaehc Patag. This river, 
unlike those which we have hitherto dealt with, 
does not flow into the Spey, but into Loch 
Laggan, after making an extraordinary volte 
face about two miles from its mouth. First it 
flows directly northwards, and then suddenly 
south-westwards for the last two miles of its 
course. Hence the local saying — 

Patag dhubh, bhalgach 
Dol an aghaidh uisge Alba 

(Dark, bubbly Pattack, that goes against the 
streams of Alba). 

We find Pattack first mentioned in an agreement 
between the Bishop of Moray and Walter Comyn 
about the year 1230, where the streams 
' ' Kyllene et Petenachy ' ' are mentioned as 
bounding the church lands of Logykenny. The 
Kyllene is still remembered in Camus-Killean, 
the bay of Killean, where the inn is. The 
Kyllene must have been the present Allt Lairig, 


or as the map has it, Allt Buidhe ; while Peten- 
achy represents Pattack, which in Blaeu's map 
appears as Potaig. The initial p proves the 
name to be of non-Gaelic origin ultimately, but 
whether it is Pictish, pre-Celtic, or a Gaelicised 
foreign Avord we cannot say. 

Ah Loiorag lies between Lochan na h-Earba 
and Loch Laggan. It means the " loud-sound- 
ing (Jahhar) one." 

The Spean; in Gaelic Spithean. See under 

We have now exhausted the leading rivers, 
but before going further we may consider the 
names of one or two tributaries of these. 
Feshie, for instance, has three important 
tributaries, one of which, Allt Mharkie, we have 
already discussed. Passing over Allt Puaidh 
as being an oblique form of Allt Ruadh, " red 
burn," we come to the curious river name 

Fernsdale; in Gaelic Fearnasdail. The 
farms of Corarnstil-more and Corarnstil-beg, 
that is, the Corrie of Fernsdale, are mentioned 
in 1603 as Corearnistaill Moir and Corearinsiail 
Beige, and in 1691 the name is CorriarnisdaiH. 
Blaeu's map gives the river as Fairnstil. ihe 
first portion of the name is easy; it is Fearna, 
alder. But what of sdail or asdailf The word 
astail means a dwelling, but "Fern-dwelling" 
is satisfactory as a name neither for river or 
glen. The tributary of the Fernsdale is called 


Comhraig; in Blaeu Conrik. Comhrag 
signifies a conflict ; but in Irish and early Gaelic 
it signified simply a meeting whether of road and 
rivers, or of men for conflict. There are 
several Irish place-names Corick, -situated near 
confluences. Doubtless this stream took its 
name from its confluence with Fernsdale. 

On Feshie we meet further up with Alii 
Fhearnagan, the stream of the alder trees; then 
AUt Gabhlach, which the Ordnance map 
etymologises into Allt Garbhlach, the stream of 
the rugged place. This may be the true deriva- 
tion; it is a big rough gully or corrie with a 
mountain torrent tumbling through it. 

Allt Lorgaidh is named after the mountain 
pass or tract which it drains {lorg, lorgadh, 
track, tracing), and which also gives name to 
the pi'ominent peak of Cam an Fhidhlei7' 
Lorgaidh, the Fiddler's Cairn of Lorgie, to 
differentiate it from the Fiddler's Cairn which 
is just beyond the Inverness-shire border, and 
not far from the other one. 

The Eidart, Blaeu' s Eitart, with the neigh- 
bouring streamlet of Eindart, is a puzzling 
name. The Gaelic is Eideard and Inndeard 
according to pronunciation. 

We now come to the lochs of Badenoch. 
Loch Alvie is bound up with the name of Alvie 
Parish, discussed already. Loch Insh is the 
Lake of the Island, just as Loch-an-eilein, in 


Rotliiemiirchus, takes its name from the castle- 
island which it contains; but eilean is the Norse 
word eyland, Eng. island, borrowed, whereas 
innis of Loch Insh is pure Gaelic. In Gaick, 
along the course of the Tromie, there are three 
lakes, about which the following rhyme is re- 
peated : — 

Tha gaoth nihor air Loch an t-Seilich 
Tha gaoth eil' air Loch an Duin; 

Ruigidh mise Loch a' Bhrodainn, 
Mu'n teid cadal air mo shuil. 

The rhyme is supposed to have been the song 
of a hunter who escaped from demons by 
stratagem and the help of a good stallion on 
whose back he leapt. The first loch is called 
Loch an t-Seilich, the lake of the willow, and 
the third of the series is Loch an Diiin, the loch 
of the Down or hill, the name of the steep crag 
on its west side. The intermediate lake is 
called Loch Vrodain, Gaelic Bhrodainn, which 
Sir S.. Gordon in Blaeu's map speUs as Vrodin. 
Ihe Ordnance map etymologises the word as 
usual, and the result is Loch Bhradainn, Salmon 
Loch; but unfortunately the a of hradan was 
never o, so that phonetically we must discard 
this derivation. There is a story told about this 
weird loch which fully explains the name 
mythically. A hunter had got into possession 
of a semi-supernatural litter of dogs. Wlien 


they reached a certain age, all of them were 
taken away by one who claimed to be the true 
owner, who left with the hunter only a single 
pup, jet black in colour, and named Brodainn. 
Before leaving it with the hunter, the demon 
broke its leg. Brodainn was therefore lame. 
There was a wonderful white fairy deer on Ben 
Alder, and the hunter decided he should make 
himself famous by the chase of it. ^ So he and 
Brodainn went to Ben Alder, on Loch Ericht 
side; the deer was roused, Brodainn pursued it, 
and was gaining ground on it when they were 
passing this loch in Gaick. In plunged the deer, 
and after it Brodainn dashed; he caught it in 
mid-lake, and they both disappeared never more 
to be seen ! Hence the name of the lake is 
Loch Vrodin; the lake is there, the name is 
there, therefore the story is true ! The word 
hrodan represents E. Ir. hrot-chu, a mastiff 
Welsh hrathgi; from brod, good, " a broth of a 
boy." Loch-Laggan takes its name from the 
lagan or hollow which gave the parish its name, 
that is, from Laggan-Chainnich or Lagan- 
Kenny, at the northern end of the loch. There 
are two isles in the lake connected with the old 
kingly race of Scotland. King Fergus, whoever 
he was, had his hunting lodge on one, called 
Eilean an Eigh, and the bther was the dog- 
kennel of these Fenian hunters, and is called 
Eilean nan Con. The considerable lake or lakes 


running parallel to, and a mile to the soiUli- 
east of Loch Laggan are called Lochan na 
h-Earba — the lakes of the roe. Loch 
Crunachan, at the mouth of Glen-Shirra, has an 
artificial island or crannog therein; the word is 
rather Crunnachan than Crunachan by pro- 
nunciation. A Gordon estate map of 1773 calls 
it the ''Loch of Sheiromore," and distinctly 
marks the crannog. Taylor and Skinner's 
Roads maps, published in 1776 by order of Par- 
liament, give the name as L. Crenackan. The 
derivation, unless referable to crannog, is doubt- 
ful. Loch Ericht, the largest lake in Badenoch, 
is known in Gaelic as Loch Eireachd. Blaeu 
calls it Eyrachle (read Eyrachte). The lake is 
doubtless named from the river Ericht, runnino- 
from it into Loch Eannoch. Another river 
Ericht flows past Blairgowrie into the Isla, nor 
must we omit the Erichdie Water and Glen 
Erichdie in Blair Athole. The word eireachd 
signifies an assembly or meeting, but there is an 
abstract noun, eireachdas, signifying " hand- 
someness," and it is to this last form that we 
should be inclined to refer the word. 

Let us now turn to the hills and hollows and 
dales of Badenoch. Many of these place-names 
are called after animals frequenting them. The 
name of the eagle for instance is exceedingly 
common in the form of iolair, as Sron an lolair, 
eagle's ness, &c. We shall begin at the north- 


east end of the district, and take the Monadh-ha 
or Grey Mountain range first. " Standing 
fast " as guard between Strathspey and 
Badenoch is the huge mass of 

Craig ellacliie, which gives its motto to the 
Clan of Grant — "Stand fast: Craigellachie !" 
The name reads in GaeHc as Eileachaidh, which 
appears to be an adjective formed from the 
stem eilech, or older ailech, a rock, nominative 
ail. The idea is the stony or craggy hill — a 
thoroughly descriptive adjective. 

The Moireach; Gaelic A' MJwr'oich, for 
A' Mhormhoich, is an upland moor of undulating 
ground above Ballinluig. On the West Coast, 
this term signifies flat land liable to sea flooding. 
It is also the real Gaelic name of Lovat. 

Cam Duhh ^Ic-an-Debir is on the Strath- 
dearn border, and is wrongly named on the map 
as " Carn Dubh aig an Doire." It means — The 
Black Cairn of the Dewar's (Pilgrim) Son. 

An Sguabach.—Theve is another Sguabach 
south of Loch Cuaicli, a few miles from Dal- 
whinnie, and a Meall an Sguabaich west of Loch 
Ericht. It means the " sweeping " one, from 
sguah, a besom. The people of Insh — the 
village and its vicinity — used to speak of the 
north wind as Gaoth na Sguabaich, for it blew 
over that hill. 

Cnoc Fraing, not Cnoc an FhrangaicJi as on 
the Ordnance map — a conspicuous dome-shaped 


hill above Dulnan river. There is a Cnoc 
Frangach a few miles south of Inverness, near 
Scaniport. Fraoch frangach means the cross- 
leaved heather, of which people ma:!e their 
scouring brushes. The brush was called in 
some parts fraings' in Gaelic. Compare M. Ir. 
f rang can, tansy. 

Easga 'n Lochain, with its caochan or 
streamlet, contains the interesting old word for 
" swamp " known as easg, easga, or easgaidh, 
with which we may compare the river name Esk. 

A' Bhuidheanaich, in the Ordnance maps 
etymologised into yi?n Buidh' aonach, " the 
yellow hill or steep," occurs three times in 
Badenoch — here behind Kincraig and Dunach- 
ton, on the north side of Loch Laggan, and on 
the confines of Badenoch a few miles south of 
Dalwhinnie. The idea of " yellowness " under- 
lies the word as it is characteristic of the places 
meant. The root is huidhe (yellow); the rest is 
mere termination and has nothing to do with 
aonach, which elsewhere is applied to a hill or 

Coire Bog, &c. — Here we may introduce a 
mnemonic rhyme detailing some features of the 
ground behind and beside Buidheanaich. 

Allt Duinne 'Choire Bhuig, 
Tuilnean agus Feithlinn, 

Coire Bog is Ruigli na h-Eag, 
Steallag is Bad-Earbag. 


The Burn of Dun-ness in Soft Corry, Dul- 
nan and Broad Bog-stream, the Eeach of the 
Notch, the Spoutie and Hinds' Clump " — that 
is the translation of the names. 

An Suidhe means the " Seat " ; it designates 
the solid, massive hill behind Kincraig. 

Craig Righ Harailt means King Harold's 
Hill, on the side of which his grave is still 
pointed out. As already said, it is unknown 
who he was or when he lived. 

Coire Neachdradh: Glac an t-Sneachdaidli , 
&c. This corrie is at the end of Dunachton buin 
after its final bend among the hills. Sneachd- 
radh means snows, or much snow — being an 
abstract noun formed from sneachd. 

Ruigh an Roig : the Eeach of the Roig (?) is 
eastward of Craig Mhor by the side of the peat 
road. The map places it further along as Ruigh 
na Ruaige — the Stretch of the Retreat. 

Bad Each is above Glen Guinack : it is mis- 
read on the Ordnance map into Pait-an-Eich — 
a meaningless expression. It means Horses* 
Clump, and a famous local song^ begins — 

Mollachd gu brath aig braigh Bad Each; 

curses ever more on upper Bad-each, where the 
horses stuck and they could not extricate them. 

1 By John Cameron, Kingussie ; migrated to America, 
where he died about 1891. 


Ehymes about the various place-names- are 
common, and here is an enumeration of the 
heights in the Monadh Liath between Kingussie 
and Craig Dhubh : — 

Creag Bheag Chinn-a'-ghiubhsaich 

Creag Mhor Bhail' a' chrothain,. 

Beinne Bhuidhe na Sroine, 

Creag an Loin aig na croitean, 

Sithean Mor Dhail a' Chaoruinn, 

Creag an Abhaig a' Bhail' -shios, 

Creag Liath a' Bhail' -shuas, 

Is Creag Dhubh Bhiallaid, 

Cadha an Fheidh Lochain Ubhaidh, 

Cadha is mollaicht' tha ann, 

Cha'n fhas fiar no fodar ann, 

Ach sochagan is dearcagan-allt, • 

Gabhar air aodainn, 

Is laosboc air a cheann. 

Glen Balloch; in Gaelic Gleann Baloch. This 
name is stymologised on the Ordnance map into 
Gleann a' Bhealaich — the Glen of the Pass; but 
the word is haloch or balloch, which means 
either speckled or high-walled. To the left the 
Allt Mhadagain discharges into the Calder : this 
name is explained on the map as Mada coin, 
which certainly is not the pronunciation which 
our Madagain reproduces. Madagan is a 
diminutive of madadh, a dog, vwlf. There are 


two corries in Gaick similarly named (Cory 
Mattakan, 1773). 

Sneachdach Slinnean, or Snow Shouldei', is 
away on the Moy border. 

Meall na h-Uinneig, behind Gask-beg con- 
siderably, means the Mass or Hill of the Win- 
dow. There are other places so named — 
Uinneag Coire an Eich (Glen-balloch), Uinneag 
Coire Ardar, Uinneag Coire an Lochain, 
Uinneag na Creig Moire, Uinneag Coire 
Chaoruinn and Uinneag Mhin Choire, the latter 
ones being all near one another on the north 
side of Loch Laggan. The meaning of the 
name is an opening or pass, or a notch in the 
sky-line. A huge cleft in a rock in the north 
end of Colonsay is called Uinneag lorcaill, 
" Hercules' Window " — Prof. Mackinnon. 

larlraig is the rising ground above Garva 
Bridge, and is mis-written for lolairig, place of 
the eagles. -"^ There is here a rock where the 
eagle nests or nested. Compare Auld Cory na 
Helrick of 1773 with the Allt Coire na h-Iolair 
of the Ordnance map, both referring to a stream 
on Loch Ericht side. There is an Elrick op- 
posite Killyhuntly. The name is common in 
North Scotland. 

Coire Yairack; AUt Yairack; in Gaelic 
Earrag, as if a feminine of Errach (spring). It 
is spelt Yarig on the 1773 estate map. Perhaps 

1 See Introduction. 


it is a corruption of Gearrag, the short one, ap- 
phed to a stream. 

Shesgnan is the name of a considerable ex- 
tent of ground near the source of the Spey, and 
it means morass land, being from seasgann, 
fenny country, a word which gives several place- 
names both in Scotland and Ireland. The most 
notable in Scotland is Shisken in Arran, a large, 
low-lying district, flat and now fertile. 

We now cross Spey, and work our way down 
the south side. 

Dearc Beinne Bige, the Dearc of the Little 
Hill. The pronunciation is dire; in the 1773 
map it is spelt Dirichk. It is an oblique case 
of dearc, a hole, cave, cleft; it is found in early 
Irish derc (a cave), and several places in Ireland 
are called Derk and Dirk therefrom. It occurs 
at least three times in Laggan — as above; and in 
Dire Craig Chathalain, the 1773 Dirichk Craig 
Caulan, or cleft of the Noisy Eock, from Callav, 
noise; and in Dearc an Fhearna. 

Coire 'Bhein, the 1773 Cory Vein, is a 
puzzling name. It looks like the genitive case 
of hian, skin. 

Coire Phitridh, at the south corner of 
Lochan na h-Earba, is given in the map as Corie 
na Peathraich. The word is probably an ab- 
stract or collective noun from pit, hollo vf. 

Beinn Eihhinn, the 1773 Bineven, the 
*' pleasant hill," is a prominent peak of 3611 


feet high, on the borders of Badenoch and Loch- 
aber, from which a good view of Skye can be 

Ben Alder, Blaeu's Bin Aildir, in modern 
Gaehc Beinn Eallar (Yallar). The word is 

Beinn Udlamaii, the Uduman of the 1773 
map, on the confines of Badenoch and Perth- 
shire, east of Loch Ericht, seems to take its 
name from the ball and socket action, for udalan 
signifies a swivel or joint. Some suggest 
iidlaidh, gloomy, retired. 

The Boar, An Tore, of Badenoch is to the 
left of the railway as one enters the district from 
the south. The "Sow of Athole " is quite 
close to the " Boar of Badenoch." We are 
now at the ridge of 

Drumochter, in Gaelic Drum-uachdar , or 
ridge of the upper ground. 

Coire Bhoite, or rather Bhoitidh, the Vottie 
of 1773, is two or three miles away, and finds 
a parallel in the name Sron Bhoitidh at the top 
of Glenfishie, where the river bends on itself. 
The word boitidh means " pig/' or rather the 
call made to a pig when its attention is desired. 

Coire Suileagach, behind Craig Ruadh and 
Drumgask, means the Corrie full of Eyes, so 
named from its springs doubtless. The term 
suileach (full of eyes) is usually applied to 
streams and corries with whirlpools therein. 


Creag Chrocan, not nan Crocean as on the 
map, is near the above corrie, and is named 
from the deer's antlers which croc means. 
Similar!}^ we often meet with cahar (an antler or 
caber) in place-names. 

The hill of Bad na Deimheis, the Bad na 
Fei^h of 1773, overlooks Dalwhinnie to the east. 
The name means the " Clump of the Shears," 
a curious designation. We now pass over into 
the forest and district of 

Gaick, in Gaelic Gaig, which is the dative or 
locative of gag, a cleft or pass. It is considered 
the wildest portion of Badenoch, and the repute 
of the district is far from good. Supernaturally, 
it has an uncanny reputation. From the days 
of the ill-starred and ill-disposed Lord Walter 
Corny n, who, in crossing at Leum na Feinne — 
the Fenian Men's Leap — to carry out his dread 
project of making the Euthven women go to the 
harvest fields to work unclothed and naked, wa:^ 
torn to pieces by eagles,^ to that last Christmas 
of last century, when Captain John Macpherson 
of Ballachroan and four others were choked to 
death by an avalanche of snow as they slept in 
that far-away bothie, Gaick has an unbroken 
record of dread supernatural doings. Duncan 

1 Henee the exi]>res&ipn — Did Blialtnir au Gaig ort — 
Walter's fate in Gaick on yoxx — to signify an ill wish or curse 
fii any one. 


Gow, in his poem on the Loss of Gaick in 1799, 
says : — 

Gaig dhubh nam feadan fiar, 

Nach robh ach na striopaich riamh, 

Na bana-bhuidsich 'gan toirt 'san lion, 

Gach fear leis 'm bu mhiannach laighe ieath'. 

Which means that Gaick, the dark, of wind- 
whisthng crooked glens, has ever been a 
strumpet and a witch, enticing to their destruc- 
tion those that loved her charms. How near 
this conception is to that mythological one of 
the beauteous maiden that entices the wayfarer 
into her castle, and turns into a savage dragon 
that devours him ! The following verses show- 
ing the respective merits of various places have 
no love for Gaick : — 

Bha mi 'm Bran, an Guile 's an Gaig, 

'N Eidird agus Leum na Larach, 
Am Feisidh mhoir bho bun gu braighe 

'S b'annsa leam bhi 'n Allt a' Bhathaich. 
'S mor a b'fhearr leam bhi 'n Drum-Uachdar 

Na bhi 'n Gaig nan creagan gruamach. 
Far am faicinn ann na h-uailsean 

'S iiibhaidh dhearg air bharr an gualain. 

The poet prefers Drumochter to Glen-Feshie and 
Gaick of the grim crags. The Loss of Gaick is 
a local epoch from which to date : an old person 



always said that he or she was so many years 
old at Call Ghaig. So in other parts, the 
Olympiads or Archons or Temple-burnings 
which made the landmarks of chronology were 
such as the " Year of the White Peas," " the 
Hot Summer " (1826?), the year of the " Great 
SnoAv," and so forth. 

" Vinegar Hill," as the maps have it, is to 
the west of the Dun of Loch an Duin ; the Gaelic 
is A' Mhin-choiseachd, the easy walking. The 
English is a fancy name founded on the Gaelic. 

A' Chaoirnich, the Caorunnach of the 
Ordnance map, but the Chournich of 1773, 
stands beside Loch an Diiin to the left. The 
latter form means the " cairny " or " rocky " 
hill; the other, the. " rowan-ny " hill, which is 
the meaning doubtless. The steep ascent of it 
from the hither end of the lake is called on the 
map Bruthach nan Spaidan, a meaningless ex- 
pression for Bruthach nan Spardan, the Hen- 
roost Brae. 

Meall Aillig, in the Gargaig Cory (1773), or 
Garbh-Ghaig (Rough Gaick as opposed to 
" Smooth " Gaick or Minigaig as in Blaeu's 
map), appears to contain aiJl (a cliff) as its root 
form. Some refer it to aileag, the hiccup, 
which the stiffness of the climb might cause. 

Coire Bhran, the Coryvren of Blaeu, takes 
its name from the river Bran, a tributary of the 
Tromie, and this last word is a well-known river 


name, applied to turbulent streams, and signifies 


Caochan a' Chaplich, a streamlet which falls 
into Tromie a little below the confluence of the 
Bran, contains the word caplach, which seems 
to be a derivative of capull (a horse). There 
is a Caiphch in the Aird — a large plateau, the 
Monadh Caiplich in Loch Alsh, and a stream of 
the name in Abernethy. 

Croyla is the prominent mountain on the left 
as one enters Glentromie— a massive, striking 
hill. It is sung of in the Ossianic poetry of 
John Clark, James Macpherson's fellow 
Badenoch man, contemporary, friend, and 
sincere imitator in poetry and literary honesty. 
Clark's (prose) poem is entitled the " Cave of 
Creyla," and in his notes he gives some 
topographical derivations. Tromie appears 
poetically as Trombia, and is explained as Trom- 
bidh, heavy water, while Badenoch itself is 
etymologised as Bha-dianach, secure valley. 
The Ordnance map renders Croyla as Cruaidh- 
leac, a form which etymologises the word out of 
all ken of the local pronunciation. Blaeu's 
map has Cromlaid, which is evidently meant for 
Croyla. The Gaelic pronunciation is Croidh-la, 
the la being pronounced as in English. It is 
possibly a form of cruadhlach or crnaidhlarh 
(rocky declivity), a locative from which might 
have been cruaidhlaigh. 


Meall an Duhh-catha is at the sources of the 
Comhraig river. It should be spelt Duhh- 
chadha, the black pass, the word cadha being 
common for pass. 

Ciste Mhairearaid or rather Ciste Mhearad, 
Margaret's kist or chest or coffin, is part of Coire 
Fhearnagan, above the farm of Achlean. Here 
snow may remain all the year round. It is said 
that Margaret, who was jilted by Mackintosh of 
Moy Hall, and who cursed his family to sterility, 
died here in her mad wanderings. 

Meall Duhhag^ and not Meall Dubh-achaidh 
(Ordnance map) is the name of the hill to the 
south of Ciste Mairead, while equally Creag 
Leathain{n), broad craig, is the name of the hill 
in front of Ciste Mairead, not Creag na 
Leacainn. Further north is 

Creag Ghiubhsachan, the craig of the fir 

Creag Mhigeachaidh stands prominently be- 
hind Feshie Bridge and Laggan-ha. There is 
a Dal-mhigeachaidh or Dalmigavie in Strath- 
dearn, a Migvie (Gaelic, Migibhidh) in Strath- 
errick, and the parish of Migvie and Tarland in 
Aberdeenshire. The root part is mig or meig, 
which means in modern Gaelic the bleating of a 

1 Meall Di(b1i-agaidh, immediateiy behind Aclilum — D. 
MoD. [Apparently a correction of Meall Dubhag ; cf. Avie- 


Creag Follais, not Creag Phulach (sic) as on 
the maps, means the conspicuous crag. 

Creag Fhiaclach, not Creag Pheacach ( !), on 
the borders of Eothiemurchus, which means the 
serrated or toothed crag, a most accurately de- 
scriptive epithet. 

Clach Mhic Cailein, on the top of Creag 
Follais. The MacCailein meant is Argyle, sup- 
posed to be Montrose's opponent, though it must 
be remembered that Argyle had also much to 
do with Huntly at Glenlivet and otherwise. 

Sgor Gaoithe (wind skerry) is behind Creag 

We have now exhausted the natural features 
of the country so far as the explanation of their 
names is necessary, and we now turn to the farm 
and field names — the hailes and townships and 
other concomitants of civilisation. Commenc- 
ing again at Craig Ellachie, we meet first after 
crossing the crioch or boundary the farm of 
Kinchyle, Cinn-Choille, wood's-end. Then 

Lynwilg, the Lambulge of 1603, LynhuiJg 
(Blaeu), signifies the land of the bag or bulge. 

BalUnluig, the town (we use this term for 
haile, which means "farm" or "township") 
of the hollow. 

Kinraim, north and south, on each side of the 
Spey. This name appears about 1338 as 
Kynroreach; 1440, as Kynrorayth; and Kynrara 


(1603). The ki7i is easy; it is ''head" or 
" end " as usual. The rara or rorath is diffi- 
cult. Rorath, like ro-dhuine (great man), 
might mean the great or noble (ro) rath or dwell- 
ing-place (the Latin villa). 

Dalraddy, Dalreadye (1603), and D(drodie 
(Biaeu). The GaeUc is Dail-radaidh, the 
radaidh dale. The adjective radaidh is in the 
older form rodaidh, which is still known in 
Gaehc in the force of " dark, sallow." A 
sallow-complexioned man might be described 
as " Duine rodaidh dorcha." The root-word is 
rod, iron scum or rusty-looking mud; it is a 
shorter form of ruadh (red). In Ireland, it is 
pretty common, and is applied to ferruginous 
land. The adjective rodaidh (dark or ruddy) 
might describe the Dalraddy land. It is in con- 
nection with Dalraddy that the great Badenoch 
conundrum is given : — 

Bha cailleach ann Dail-radaidh 
'S dh' ith i adag 's i marbh. 

(There was a carlin in Dalraddy who ate a had- 
dock, being dead). With Dalraddy estate are 
mentioned in 1691 the lands of Keanintachair 
(now or lately Cinn-tachair, causeway-end), 
Knockningalliach (the knowe of the carlins), 
Loyninriach, Balivuilin (mill-town), and the 
pasturages Feavorar (the lord's moss-stream), 
Riochnabegg or Biachnabegg, and Batabog (now 


Bata-bog, above Ballinluig, the soft swampy- 
place). Another old name is Gortincreif 
(1603), the govt or field (farm) of trees. CVv/i- 
gowan means the Smith's Croft. 

DelfouT, DalpJiour in 1603, and older forms 
are Dallefowr (1569). The del or dal is for dale, 
but what is four? The Gaelic sound is fur. 
The word is very common in names in Pittland, 
such as Dochfour, Pitfour, Balfour, Letterfoiir, 
Tillyfour, Tillipourie and Trinafour. These 
forms point to a nominative piir, the p of which 
declares it of non-Gaelic origin. The term is 
clearly Pictish. The only Welsh word that can 
be compared is paivr (pasture), pori (to graze), 
the Breton peur. Fiir ha.s nothing to do with 
Gaelic fuar, for then Dalfour would in Gaelic 
be Dail-fhuar, that is Dal-uar. 

Pitchurn, in 1603 Pettechaerne, in Gaelic 
Bail-chaorruinn, the town of the rowan. The 
Pictish pet or pit (town, farm), which is 
etymologically represented by the Gaelic cuid, 
has been changed in modern Gaelic to haile, the 
true native word. 

Pitourie, in 1495 Pitwery, in 1603 
Pettourye, in 1620 Pettevre, &c. ; now 
BaiVodharaidh. The adjective odhar means 
" dun," and odharach, with an old genitive 
odharaigh, or rather odharach-mhullach, is the 
plant devil's bit. The plant may have given 
the name to the farm. 


Baldoxc means the black town. 

Kincraig, Kyncragye (1603), means the end 
of the crag or hill, which exactly describes it. 

Leault, Gaelic Leth-allt or half-bm'n, a name 
which also appears in Skye as Lealt, may have 
reference rather to the old force of allt, which 
was a glen or shore. The stream and partly 
one-sided glen are characteristic of the present 

Dunachton; Gaehc Dun-Neachdain{n), the 
hill-fort of Nechtan. Who he was, we do not 
know. The name appears first in history in con- 
nection with the Wolf of Badenoch. St Dros- 
tan's chapel, below Dunachton House, is the 
capeUa de Nachtan of 1380. We have 
Dwnachtan in 1381, and Dunachtane in 1603. 
The barony of Dunachton of old belonged to a 
family called MacNiven, which ended in the 
15th century in two heiresses, one of whom, 
Isobel, married William Mackintosh, cousin of 
the chief, and afterwards himself chief of the 
Clan Mackintosh. Isobel died shortly after 
marriage childless. Tradition says she was 
drowned in Loch Insh three weeks after her 
marriage by wicked kinsfolk.-^ Mr Fraser-Mac- 
kintosh has written a most interesting mono- 
graph on Dunachton, entitled " Dunachton, 
Past and Present." 

1 1475, Baron Macknenan deceased; Laclilan of Gr«llovy 
gets the marriage of his daughters [no reference.] 


Achnaheachin; Gaelic Ach' nam Beath- 
aichean, the lield of the beasts. Last century 
this land held eight tenants. 

Keppoclunuir; Gaelic An Sliahh Ceapanach; 
Ceapach means a tillage plot. 

Coilintuie or Meadoivside. The Gaelic is 
Coin an t-Suidke, the Wood of the Suidh, or 
sitting or resting. Some hold the name is really 
Cuil an t-Shuidh, the Recess of the Suidh. 

Croftcarnoch; Gaelic Croit-charnach, the 
Cairny Croft. 

Belleville is, in its English form, of French 
origin, and means " beautiful town." The old 
name in documents and in maps was Raitts, and 
in the 1776 Roads' Map this name is placed 
exactly where Belleville would now be written. 
Gaehc people call it Bail' a' Bhile, " the town 
of the brae-top," an exact description of the 
situation. Mrs Grant of Laggan (in 1796) says 
that Bellavill ' ' is the true Highland name of the 
place," not Belleville; and it has been main- 
tained by old people that the place was called 
Bail' a' Bhile before " Ossian " Macpherson 
ever bought it or lived there. Whether the 
name is adopted from Gaelic to suit a French 
idea, or vice versa, is a matter of some doubt, 
though we are inclined to believe that James 
Macpherson was the first to call old Raitts 
by such a name. James Macpherson 
is the most famous — or rather the most 


notorious — of Badenoch's sons; bjit though 
his " Ossian "is a forgery from a historical 
standpoint, and a purely original work from a 
hterary point of view, yet it is to him that Celtic 
literature owes its two greatest benefits — its 
being brought prominently before the European 
world, and, especially, the preservation of the 
old literature of the Gael as presented in tradi- 
tional ballads and poems, and in the obscure 
Gaelic manuscripts which were fast disappear- 
ing through ignorance and carelessness. 

Lachandhu, the httle loch below Belleville, 
gives the name to Sir Thomas Dick Lauder's 

Raitts — the Enghsh plural being used to de- 
note that there were three Raitts — Easter, 
Middle, and Wester. In 1603 the ^lace is 
caUed Reatt, and Blaeu has Rait. The Gaehc is 
Rat, and this, which is the usual form in High- 
land place-names, is a strengthened form of the 
older rath or rdith of Old Irish, which meant a 
residence surrounded by an earthern rampart. 
It, in fact, meant the old farm house as it had 
to be built for protective purposes. For the 
form rat (from rath-d), compare Bialaid, further 
on, and the Irish names Kealid from caoJ and 
Croaghat from cruach, which Dr Joyce gives in 
his second volume of Irish Place-Names to ex- 
emplify this termination in d. 


Chapel-park ; Gaelic Pairc an t-Seipeil. This 
is a modern name, derived from the chapel and 
kirk-yard that once were there, which was 
known as the chapel of Ma Luac, the Irish Saint. 
The older name was the Tillie or Tillie-sow, 
where an inn existed, whose " Guidwife " was 
called Bean an Tillie. Some explain Tillie-sow 
as the Gaelic motto that used, it is said, to be over 
the olden inn doors, viz., " Tadhaihbh so " — 
"Visit here." 

Lynchat is now BaiV a' Chait; " Sheep-cot " 
town, not Cat's-town, is the explanation given 
by the inhabitants. [So also Aodann Chat, 
Edincat, in Strathdearn.] 

An Uaimh Mhoir, the Great Cave, is a quar- 
ter of a mile away from the highway as we pass 
Lynchat. It is an " Erd-house," the only one 
of this class of antiquarian remains that exists 
in Badenoch. It is in the form of a horse-shoe, 
which has one limb truncated, about 70 fett 
long, 8 feet broad, and 7 high. The walls 
gradually contract as they rise, and the roofmg 
is formed by large slabs thrown over the ap- 
proaching walls. Tradition says it was made m 
one night by a rather gigantic race : the women 
carried the excavated stuff in their aprons and 
threw it in the Spey, while the men brought llif 
stones, large and small, on their shoulders from 
the neighbouring hills. All was finished by 
morning, and the inhabitants knew not what had 


taken place From this mythic ground we come 
down to the romantic period, when, according 
to the legend, MacNiven or Mac Gille-naoimh 
and his nine sons were compelled to take refuge 
here — some say they made the cave, and long 
they eluded their Macpherson foes. There was 
a hut built over the mouth of the cave, and at 
last it was suspected that something was wrong 
with this hut. So one of the Macphersons 
donned beggar's raiment, called at the hut, pre- 
tended to be taken suddenly ill, and was, with 
much demur, allowed to stay all night. There 
was only one woman in the hut, and she was con- 
tinually baking; and he could not understand 
how the bread disappeared in the apparent press 
into which she put it, and which was really the 
entry into the cave. He at last suspected the 
truth, returned with a company of men next 
night, and slew the MacNivens. It is said that 
this man's descendants suffered from the ailment 
which he pretended to have on that fateful night. 

Laggan, the hollow, now in ruins. Here 
dwelt the famous Badenoch witch, Bean an 

Kerroiv ; in Gaelic An Ceathramh, the fourth 
part — of the davoch doubtless — the davoch of 
"Kingussie Beige" (1603), with its "four 

Kingussie. Already discussed under th'^ 
heading of Kingussie parish. 


Ardvroilach; Gaelic Ard-hhroighleach; in 
1603, Ardhrelache . The form broighleacli 
seems a genitive plural from the same root form 
as hroighleag, the whortleberry. The word 
hroighlich (brawling) scarcely suits with ard a 

Pitmain. The Gaelic is only a rendering of 
the Enghsh sounds : Piodme'an. In 1603 it is 
Petmeane. The reason for their being no Gaelic 
form of this word is simply this. The gre;it 
inn and stables of the Inverness road were here, 
and the name Pit-meadhan, " middle town,''' 
was adopted into the English tongue. The 
Gaelic people, meantime, had been abolishing 
all the pet or pit names, and changing them lo 
Bals, but this one was stereotyped in the other 
tongue, and the local Gael had to accept the 
English name or perpetuate an offending form. 
He chose to adopt the English pronunciation. 

Balachroan; Bellochroan (1603); Gaelic 
Baile-'Chrothain, the town of the sheepfold. 
Above it was Coulinlinn, the nook of the lint, 
where an old branch of Macphersons lived. 

Aldlarie; Gaelic Allt-Lairigh, the stream of 
the larach or gorge. 

Strone means " nose." 

Neujtonmore is the new town of the Moor — 
An SHabh. 


Clune and Craggan of Clune. The Gaelic 
chiain signifies meadow land, whether high or 
low, in dale or on hill. 

Benchar, Bannachar (1603), Beandocher 
(1614), and now Beannachar, Irish heannchar 
(horns, gables, peaks), Welsh Bangor. It is a 
very common place-name. The root is heann 
or heinn (a peak). 

BeaUid, in 1603 Ballet, in 1637 Ballid, now 
Bialakl, so named from being at the mouth of 
Glen-banchor — hial (mouth), with a termination 
which is explained under Raitts. A 
" pendicle " of it, called Corranach, is often 
mentioned, which probably means the 
" knowey " place. 

Cladh Bhnd and Cladh Eadail, Bridget's 
and Peter's (?) Kirk-yards, are the one at Ben- 
char and the other along from Beallid, the latter 
being generally called Cladh Bhiallaid. Chapels 
existed there also at one time. 

Ovie, in 1603 Owey (and Corealdye, now 
Coraldie, corrie of streams or cliffs), Blaeu's 
Owie, now Uhhaidh, appears to be a derivative 
of uhh, egg : it is a genitive or locative of 
uhhach, spelt and pronounced of old as uhhaigh. 
Mrs Grant describes Lochan Ovie as beauty in 
the lap of terror, thus suggesting the derivation 
usually given of the name, viz., uamhaidh, 
dreadful. Some lonesome lakes of dread near 


Ballintian are called Na h-uath Lochan, the 
dread lakes. 

Cluny, Clovnye (1603), now Cluainidh. The 
root is cluain (meadow), and the termination is 
doubtless that in A' Chluanach, a cultivated 
plateau behind Dunachton, and the dative sin- 
gular of this abstract form would give the 
modern Cluny from the older cluanaigh. 

Balgowan, Pettegovan (1603) now BaiV a 
Ghohhainn, the town of the smith. 

Gask-heg, Gask-more, Gargask, Drumgask 
— all with Gask, and all near one another about 
Laggan Bridge. There is an older Gasklone, 
Mud-Gask, the Gascoloyne of 1603, Gasklyne 
(1644), and Gaskloan (1691). The form Gask 
appears in the Huntly rental of 1603. The 
name Gask is common; there is Gask parish in 
Strathearn, Perthshire, and there is a Gask in 
Strathnairn, a Gask Hill in Fife, and Gask House 
near Turriff. The name Gaskan appears more 
than once, and in one instance applies to a rushy 
hollow (Gairloch). We have Fingask in four 
counties — Aberdeen, Fife, Inverness (in the 
Aird, but the Gaelic is now Fionn-uisg'), and 
Perth. Colonel Eobertson, in his " Topo- 
graphy of Scotland," refers Gask to gasag, 
diminutive of gas, branch; but this hardly suits 
either phonetically or otherwise. The word 
gasg seems to have slipped out of use : it belongs 
only to Scotch Gaelic, and may be a Pictish 


word. The dictionaries render it by " tail," 
following Shaw, and mis-improving the matter 
by the additional synonym ' ' appendage , ' ' 
which is not the meaning ; for the idea is rather 
the posterior of an animal, such as that of the 
hind, which Duncan Ban refers to in this case 
as "white" — " gasganan geala," and which 
makes an excellent mark for the deer-stalker. 
The dictionaries give gasgan, a puppy; 
gasganach, petulant; and gasgara (gasganaf), 
posteriors; all which Shaw first gives. There 
is also the hving word gasgag, a stride, which no 
dictionary gives. These derivations throw very 
little hght on the root word gasg, which seems 
to signify a nook, gusset, or hollo v;. The 
Laggan gasgs are now ' ' rich meadows, bay 
shaped," as a native well describes them. it 
was at Gaskbeg that the gifted Mrs Grant of 
Laggan lived, and here she sang of the beauties 
of the Bronnach stream — the Gaehc Bronach, 
the "pebbly" (?) — which flows through the 

Blargie, in 1603 Blairovey, in Blaeu Blariki, 
and in present Gaehc Blaragaidh. The ter- 
mination agaidh appears also in Gallovie, which, 
in 1497, is Galoivye, and now Geal-agaidh, the 
white agaidh. The word appears as a prefix m 
Aviemore and Avielochan, both being agaidh 'n 
Gaelic. The old spehing of these words with a 
r, as against the present pronunciation with g, 


is very extraordinary. The meaning and 
etymology of agaidh are doubtful. Shaw gives 
aga as the " bottom of any depth," and there is 
a Welsh word ag, a " cleft or opening." The 
word may be Pictish. 

Coull, in Gaelic Cuil, means the " nook, 
corner," which the place is. 

Ballmishag means the town of the kid, 
mlseag or minnseag. 

Crathie, in 1603 Crathe, in Blaeu Crachy, 
now in Gaelic Craichidh. The name appears in 
the Aberdeenshire parish of Crathie, pro- 
nounced by Gaelic natives as Creychie ; Creychin 
in 1366. Crathienaird is in 1451, Crachenardy. 
[Cray at foot of Glenshee is in Gaelic Crathaigh; 
Loch Achray is in Gaelic Loch Ath-chrathaigh : 
? Ach-ch.] The form Crathie possibly points 
to an older Gaelic Crathigh. 

Garvaheg and Garvamore, the Garvey Beige 
and Garvey Moir of 1603. The word at pre- 
sent sounds as Garhhath, which is usually ex- 
plained as garhh-ath, rough ford, a very suitable 
meaning and a possibly correct derivation. 

Shirramore and Shirraheg, the Waster 
Schyroche and Ester Schiroche of 1603. 
Sheiro-more, in 1773, is in Gai,e\ic Siorrath Mor. 
With these names we must connect the adjoin- 
ing glen name, Glenshirra, Gaelic Glenn Sioro, 

a name which appears also in Argyleshire, near 



Inveraray, as Glenshira, Glenshyro (1572), tra- 
versed by the Shira stream. The root word ap- 
pears to be sh' or sior, long. Some suggest 
siaradh, squinting, obhqueness. 

Aherarder, Blaeu's Ahirairdour , Gaelic 
Obair-ardur. There is an Aberarder (Aberar- 
dor in 1456, and Abirardour in 1G02) in Strath- 
nairn, and another in Deeside, and an Auchter- 
arder in Strathearn. The Aber is the Pictish 
and Welsh prefix for " confluence," Gaelic 
inver. The ardour is etymologised in the Ord- 
nance map as Ard-dhoire, high grove. The 
word may be from ard dhohhar, high water, for 
the latter form generally appears in place-names 
as dour. 

Ardverikie has been explained correctly in 
the "Province of Moray," published in 1798, 
as ' ' Ard Merigie, the height (for rearing the 
standard." The Gaehc is Ard Mheirgidh, from 
meirge, a standard. 

Gallovie.- — See under Blargie. 

Muccoul is from Muc-cuil, Pigs' nook, 

Ach-duchil means the field of the black 

Dalchully, Gaelic Dail-chuilidh. The word 
cuilidh signifies a press or hollow. It means the 
" dale of the hollow or recess." 

Tynrich is for Tigh an Fhraoich, house of 
the heath. 


Catlodge, in 16U3 Catteleitt, and in 1776 
Catleak, is in present Gaelic Caitleag, the Cat's 
Hollow; some suggest cat, sheepcote. The 
form cait is unusual; we should, by analogy with 
Muc-ciiil and other names where an animal's 
name comes first in a possessive way, expect 
Catlaig rather than Caitleag. 

Breakachy, Brackachye (1603), is usually 
explained as Breacachaidh, speckled field, which 
is correct. " Brecacath " (Monymusk) is ex- 
plained in the 16th century as " campus 
distinctus coloribus." [Compare Ardochy, 
Highfield, in Stratherrick, Strathdearn and 
Lome.] We shall now cross the hills into Glen- 
truim and up Loch Ericht side. There at Loch 
Ericht Lodge we have 

Dail an Longairt, in 1773 Rea DeIe7ilongarf, 
and on the other side of the ridge is Coire an 
Longairt (Cory Longart 1773), while there is ari 
Eilean Longart above Garvamore bridge and 
'" Sheals of Badenlongart " in Gaick above the 
confluence of Bran, according to the 1773 map. 
Longart itself means a shealing, the older form 
being longphort, a harbour or encampment. 

Dalwhinnie, in Gaelic Dail-chuinnidh, is 
usually explained as Dail-choin7iimh, Meeting's 
Dell; but the phonetics forbid the derivation. 
Professor Mackinnon has suggested the alterna- 
tive of the " narrow daiL" Dalwhinnie was a 
famous station in the old coaching days, and the 


following verse shows how progress northwar " 
might be made : — 

Bracbhaist am Baile-chloichridh 
Lunch an Dail na Ceardaich 
Dinneir an Dhail-chuinnidh 
'S a' bhanais ann an Rat. 

Presmitckerach, not the Ordnance Pres- 
mocachie, is in 1603 Presmukra, that is Preas- 
mucraigh, bush of piggery or pigs. 

Dalannach, which the Ordnance map etymo- 
logises into Dail-gleannach or Glen-dale, was in 
1603 Dallandache, and is now Dail-annach. The 
old form points to the word lann or land, an 
enclosure or glade. The Irish Armagh, for 
Eanach, a marsh, will scarcely do, as the name 
appears in Loch Ennich in its proper Gaelic 

Cruhinmore, Crohine (1603), now Criihinn. 
The names Cruheen, Cruhoge, Slievecrooh, &c., 
appear in Ireland, and are referred by Dr Joyce 
to cruh, (a paw, hoof), criiihm (a trotter, little 
hoof). The Gaelic cruhach (lame), and cruhan 
(a crouching), are further forms of the root 
word, a locative case from the the latter form 
being possibly our Crubin, referring to the two 
" much back-bent hills there." 

Invernahavon, Invernavine (1603), means 
the confluence of the river, that is, of the Truim 
witli Spey. 


Ralia, Gaelic Rath-liath, means the grey 
rath or dwelling-place. 

Nuide, Nuid (1603), Noid (1699), now Noid. 
The derivation suggested for the name is nuadh- 
id, a topographic noun from the adjective nuadh 
or nodha, new; of old, " Noid of Ealia." 

Knappach, in Gaelic A' Chnapaich, the hilly 
or knobby land. It is a common place-name, 
especially in Ireland, appearing there as Knap- 
pagh and Nappagh. 

Ruthven, which is also the first form the 
name appears in in 1370, when the "Wolf" 
took possession of the lordship of Badenoch. It 
was here he had his castle. In 1380 the name 
is Rothven and Ruthan. The name is common 
all over Pictland, mostly in the form Ruthven, 
but also at various times and places spelt Rutii- 
fen, Ruwen, Ruven, Riv(v) en, &c. The 
modern Gaelic is Ruadhainn, which simply 
means the " red place," from ruadhan, any- 
thing red. The v of the English form lacks 
historic explanation. Brae-ruthven gives the 
phonetically interesting Gaelic Bre-ruadhnach. 

Gordon Hall (so in 1773 also) is in Gaelic 
Lag an Notair, the Notary's Hollow, for it is a 
hollow. The name and its proximity to Ruth- 
ven Castle mutually explain one another : Gor- 
don Hall was doubtless the seat of the Gordon 
lords of Badenoch, when the castle of Ruthven 
was changed to barrack purposes. Here the 


rents used to be " lifted ' ' for the Gordon 

KiUiehuntlij, KeiUehuntlye (1603), Blaeu's 
Kyllehunteme , in present Gaelic Coille-Chun- 
tainn, the wood of Contin. Himtly is in Gaelic 
Hundaidh, and M'Firbis, in the 16th century, 
has Hundon; hence arises the English form. 
The popular mind still connects it with the 
Huntlies. Contin is a parish in Ross-shire, and 
there was a Contuinn in Ireland, on the borders 
of Meath and Cavan, which is mentioned in con- 
nection with Fionn's youthful exploits. It has 
been explained as the meeting of the waters, 
con- (with) and tuinn (waves), but the matter is 

Inveruglas, In7ieruglas (1603), in Gaelic 
Inhhir-ulais, the inver of Ulas, although no such 
stream exists now, receives its explanation from 
the old Retours, for in 1691 we have mention of 
Inveruslash and its mill-town on the water of 
Duglass, which means the stream passing the 
present Milton. Hence it means the inver of 
Duglass or dark stream, dahh (black), and glais 

Soillierie, in Gaelic Soileiridh, means the 
"bright conspicuous place," on the rising be- 
yond the Insh village. 

Lynchlaggan stands for the Gaelic Loinn- 
Chlaiginn, the Glade of the Skull, possibly refer- 
ring to the knoll above it rather than to an 


actual skull there found ; the name is applied in 
Ireland to such skull-like hills. 

Am Beithigh (not Am Beithe), means the 
Bnch-pool; a locative form. 

Farletter is the old name for Balnacraig and 
Lynchlaggan, and it appears in 1603 as Ferlatt 
and Falatrie (1691). It took its name from the 
hill above, now called Craig Farleitir. The 
word Farleitir contains leitir, a slope or hillside, 
and possibly the preposition for (over), though 
we must remember the Fodderletter of Strath- 
avon with its Pictish Fotter, or Fetter, or 
Father (?). 

Forr is situated on a knolly ridge overlook- 
ing Loch Insh, and evidently contains the pre- 
position for (over), as in orra for forr a, on them. 
The last r or ra is more doubtful. Farr, in 
Strathnairn and Sutherland, is to be compared 
with it. 

DaJnavert, in 1338 and 1440 Dalnafert, in 
1603 Dallavertt, now in Gaelic Dail-a'-hheirt, 
which is for Dail na hhfeart, the dale of the 
graves or trenches, from feart, a grave, which 
gives many place-names in Ireland, such as 
Clonfert, Moyarty, &c. 

Cromaran is possibly for Crom-raon, the 
crooked field. 

Balnain is for BeaJ an athain, the ford 


Ballintian, the town of the fairy knoll, was 
called of old Countelawe (1603) and Cuntelait 
(1691), remembered still vaguely as the name 
of the stretch up the river from Ballintian, and 
caplained as Cunntadh-laid, the counting (place) 
of the loads ! Perhaps, like Contin, it is for 
Con-tuil-aid, the meeting of the waters, that is, 
of Feshie and Fernsdale, which takes place 

Balanscrittan, the town of the sgriodan or 
running gravel. 

Bulroy, for Bhuaile-ruaidh, the red fold. 

Tolvah, the hole of drowning. 

Achlean, for Achadh-leathainn, is broad 
field. Beside it is Achlum, for Achadh-leum, 
the field of the leap. 

Ruigh-aiteachain may possibly be a corrup- 
tion for Ruigh Aitneachain, the Stretch of the 

Ruigh-fionntaig , the Reach of the Fair- 

In the Dulnan valley is Caggan, the Gaelic 
of which is An Caiginn, and there is " a stony 
hill face " in Glen-Feshie of like name. 



Much mystery is made to attach to this name, 
though, as a matter of fact, the word is simple 
enough as to meaning. It is obsolete both m 
Scotch and Irish Gaelic, and it is usually glossed 
by eacilais (church) simply. There was, how- 
ever, a dilference between the general term 
eaglais and the restricted word ayinoid. The 
annoid church was that in which the patron saint 
(of the monastery or monastic district) was edu- 
cated 01 in which his relics were kept (i mhi 
taisi in erloma). The first time we meet with it 
in literature is in the Book of Armagh (circa 800 
A.D.), and there Iserninus, or larnan, is left at a 
certain spot by St Patric to found his monastery 
(rnanche) and his patron saint's church (andooit). 
The Old Irish form is andoit, with the last vowel 
long; and its derivation is disputed. Dr Whitley 
Stokes suggests antitas, antiquity (" ancient 
church "), as its origin, a late Latin term, the 

284 ANN AT. 

phonetics being the same very much as in 
Trianaid, O.I. Trmdoit. Unfortunately antitas 
is a figment of the philological brain to explam 
a7itas (G. antatis), which is glossed by 
" senatus " or senate, its meaning bemg practi- 
cally that of ecclesia, which became the Gaelic 
eaglais. The phonetics here are exactly those 
of Trianaid. We may take it that mediaeval 
Latin antas, from ante, before, and really mean- 
ing " council of ancients," is the ancestor of 

The relation of the anndid to other churchea 
in the district was one of superiority and 
antiquity. It is especially contrasted with the 
dalta church (cf . Kildalton), founded by a mem- 
ber of the same community as the founder of 
the original church and monastery, and the word 
may be translated as " sister " or " fosterling " 
church. A further church, another step below 
it, was the conpairche (co-parishioner) church 
under the tutelage of the original founder. Two 
words are general, eaglais and ceall or cell, now 
known only by its locative cill (Kil-), the latter 
meaning a smaller church than the other usually, 
being from Lat. cella, a cell. Other church 
names meet us having shghtly different mean- 
ings : — teampull, church, originally ' ' temple ' ' ; 
clachan, church, Irish clochdn, a stone bee-hive 
monastic hut; seipeal, Middle Irish sepell, a late 
word from chapel; neimheadh, glebe, Old Irish 

ANNAT. 285 

nemed, chapel, Gaulish nemeton. This last is 
the original Celtic name for a temple, and comes 
from the same root as neamh. It appears rarely 
— Eosneath (old Neveth), Navity once or twice, 
and especially m Eoskeen, where Mr Watson in 
his " Eoss-shire Place-Names" has shown the 
word to exist in Nonakil (neimh' na cill'), Dal- 
navie, Knocknavie, Inshnavie, and Newmore 
(neimh' mhor). Annoin or Andoin (church) is 
another name that only appears in the glosses. 

Annaid is therefore not a native word. Its 
chanee resemblance to the eastern goddess 
Anaitis has been responsible for much '' Druid- 
ism," and bad speculation on Celtic religion 
generally. In this connection we may mention 
two other interesting ecclesiastical words. The 
first is Manachainn or Monastery, the Gaelic 
name of Beauly. The other is the early Irish 
apdaine or ahthaine, abbacy, or " abbey land " 
also, whence our two or three Appins, about 
which also much nonsense is usually written. 
As ahthane, a supposed title, the word puzzled 
the historians for many centuries, until Skene, 
himself first a victim, discovered the mistake. 
In proof of the above facts in regard to annaid, 
the many glossaries of early Irish published with 
Irish texts must be consulted, but a fair account 
of the matter can be got from the third volume 
of the published Senchus Mor. 





The new Saga book of the Viking Club discusses 
the origin of ark and erg in the place-names of 
northern England, and tries to overturn the 
theory that they are from Norse horgr {Horg)^ 
and Anglo-Saxon hearg, a sacrificial " grove " 
of heathen times. The new theory regards them 
as being from Norse or Danish erg or cerjj, a 
shieling or dairy farm, a word undoubtedly bor- 
rowed by the Norse as the Orkney Saga fully 
proves, and as several place-names in the High- 
lands and Isles still prove. Dr Colley-March 
was the originator of the new theory in a paper 
printed in 1890 in a Liverpool antiquarian 
society's transactions — and I have not seen it; 
but as Dr Isaac Taylor in his excellent work on 
" Names and their Histories " (1896), holds by 



the horg theory, Dr March's view is either im- 
known to or rejected by the Enghsh experts on 
place-names. The difficulties in both theoiies 
are great : horg can hardly be used with other 
than a god's' name outside epithets; of course it 
is used alone in Harrow. It is difficult to equate 
Grims-argh in Preston with a deity. Again the 
borrowed arg of the Norse cannot without great 
difficulty be connected historically with northern 
England. In the Highlands the termination ary 
in place names is common, less so is sary; the 
latter nearly always comes from the possessive s 
before arj/, and in the Norse drg ; the former 
may belong to other endings, especially -gcLTvy 
(N. gerdi, G. gearraidh, outland beyond town- 
ship ploughed land). The only literary refer- 
ence in Norse to arg or erg is in the Orkney Saga, 
where we have the place called by them Asgrims- 
^rgin practically glossed by the expression 
'■ erg, which we call setr (sheihng)." Asgrims- 
serg is now called Askary or Assary, at the north 
end of Loch Calder in Caithness. When one 
compares the original form Asgrims-serg with 
the present Askary or Assary, one is compelled 
to tremble (metaphorically) for the etymologist 
of Western Isles names of Norse origin. Pr 
Anderson points out that many places in Caith- 
ness present this termination — Halsary, (Hall, 
or. perhaps, Hallvard !), Dorrery, Shurrery, 


Blingery, &c. Sutherland presents at least 
three — Gearnsary, Modsary, and Gradsary, but 
with Asgnms-£erg before our eyes, we refuse at 
present to consider them, though Mr Mackay of 
Hereford has made a decent attempt to etymolo- 
gise them in Vols. XVII . and XVIII. of Inverness 
Gaelic Society Transactions. To regain con- 
fidence, we must go to the happy shelling 
grounds of -sary and -ary in the Uists. In North 
Uist we have two distinct districts given over to 
Aulasary, which, of course, is Olafs-arge {arge 
must have been the oldest form, as we shall see), 
and which means " Olave's Shelling." In the 
same island is Obisary, which stands for Hops- 
arge, " Sheihng in the Bay." There, too, we 
have Langary from lang, long; Eisary, from 
Hris, copse; Horisary (horgs, "grove"?), 
Dusary, Vanisary, and Honary. In South Uist 
are Vaccasary and Trasary, in Barra is Ersary 
(Eric's-arge?). Ardnamurchan seems to con- 
tain, some: Brunery {brannr, spring), Smirisar- 
ary (" smear or butter " ?), Ahsary and Assary, 
in Glenelg Skiary. But these last five I do not 
know the pronunciation of, and may not be 
rightly included. Perhaps some native may 
oblige on this point. The Gaelic airigh, mis- 
spelt airidh, is in early Irish airge, dairy or a 
place where cows are, which in old Irish would 
be arge, at which stage the Norse borrowed it 


from the Scots. Personally I believe that it was 
adopted only in the Highlands by them. ^^ 
the bye, its initial use has been suggested ror 
Arkle in Sutherland, that is Arg-fell, " Sheil- 
ing's Fell " ; if so, the difficult ar or ark of Arbol, 
in Easter Ross, might so be explained. The 
English forms from ar^ • generally show ark, if 
the root is initial in the word. In future it is 
hoped that any Gaelic writer who reads the above 
will write airigh, not airidh, for " shelling." 



The glen is, as usual in such cases, named after 
the river Shiel. There is another river Shiel 
and Loch Shiel forming the southern boundary 
of Moidart. It is mentioned by the Dean of 
Lismore, who gives the bounds of the Clanranald 
as — 

Eddir selli is sowyrrni. 
" Between Shiel and Loch-hourn (Sorn)." 

Fortunately, we can go back eight hundred years 
further than the Dean's time, and we find the 
southern Shiel in Adamnan twice under the form 
of Sale. The phonetics here are all right, for 
Old Irish saile, sahva, is in modern Gaelic sile. 
The old Celtic form of the river's name would be 
Salia, the root being sal, sea, salt, sahva. The 
root s-val, swell, is also possible. The name 
hkely is Pictish, and therefore we are forced to 
fall back on original Celtic roots for an explana- 



Your correspondent, " Clach," has forgotten 
the late " Clach's " (Alex. Mackenzie) deriva- 
tion of Tomnahuirich ; this was Tom-na-Fiodhr- 
aich, " Hill of the Wood." In favour of this 
he quoted Thomas Mackenzie, headmaster of 
Eaining's School, and after 1843 connected with 
the High School. Old Thomas declared the 
" f " was dropped within the memory of people 
living in his time, and the old cailleachs of the 
town used to go out to get firewood there, speak- 
ing of it as " dol an fhiodhrach." But this is 
fanciful. The word fiodhrach does not mean 
wood " in the sense of trees, but of logs for 
ehipbuilding ; even so it is rare. Again, we have 
the Gaehc pronunciation of Tomnahurich re- 
corded in the Fernaig MS. (1690) in the " Pro- 
phesie about Inverness." "There will be 
battle — 

i dig McPehaig i mach 

Lea layn agus lea luhrich 

Tuitti ni Ghayle ma saigh 

Ma voirlumb toim ni hurich 


— ill wliich Macbeth will come forth with sword 
and armour, and the Gael will fall over other on 
the Bordland of Tom-na-hurich." Mr Thomas 
Mackenzie's derivation is simply impossible, and 
it is grammatically bad. Ballifeary no doubt 
means " Town of the Watch," but I know no 
Gaelic words that could make Tomnahurich into 
"Watchman's Hill." A very usual derivation 
has been the "Boat Hill," from its turned up 
boat shape, and with this fancy the cemetery at 
the top seems to have been made into a sort of 
ship's shape. The name iuhhrach for boat is 
poetical, and derived from the name of the fatal 
mythic vessel that conveyed Clan Uisneach back . 
to Ireland. The name simply means the " Yew 
Ship." For, after all wanderings, we must fall 
back on the manifest meaning. Tom-iia- 
h-Iubhraich means simply " Hill of the Yew 
Wood"; iuhhrach means a yew wood, from 
iuhhar, yew, just as giuthsach means " Pine- 
wood " (Kin-gussie is Cinn-ghiuthsaich, older 
ghiusaigh); beitheach, birch-wood, and so on. 
The old Irish for iuhhar was ihar and the Gaulish 
etem is ehuro, common in place names and even 
tribal names. York was called Eburacum, 
which is much the same form as iuhhrach, which 
stands for Celtic Ehurdkon. 




Place-Names of Ross and Cromarty. By W. 
J. Watson, M.A. (Aberd.), B A. (Ox.ou ). 
Inverness : Northern Counties Printing and 
Publishino- Co. 1904. 10s 6d. 

Mr Watson's book on the place-names of Ross 
and Cromarty holds a unique position : it is the 
first attempt by a Gaehc-speaking Celt, trained 
in modern philologic ways, to give in book-form 
the results of a thorough investigation into the 
names of a large county, and, incidentally, to 
give a practical epitome of Scottish place-names. 
Many years ago — in 1887 — -Professor Mac- 
kinnon published in the Scotsman a series of 
articles on " Place and Personal Names in 
Argyle," marked 'by that modern scholarship 
which native Gaelic speakers so abundantly lack 
in dealing with such matters, but, unfor- 
tunately for the public, he has never gathered 
them into book-form. There have also been 
several other competent, and, at the same time, 
Gaelic-speaking philologists who dealt with the 
place-names of different localities in papers and 
articles more or less fugitive. Mr Watson, 
however, is really the first Gael in the field with 
a work which can be honestly called scientific, 
which systematises its results in a way helpful 


for investigators in this diflicult subject. It has 
only been too painfully evident of late years that 
only a learned native Gael — or a German ! — can 
really deal with the Celtic names of Scotland. 
Hitherto the authors of works on Scottish place- 
names have not taken the trouble to learn the 
Gaelic language — and that, too, a language 
which possesses a double set of inflections, 
initial, and, as usual, final, not to mention the 
fact of its difficult Continental pronunciation. 
It is no wonder that one Sassenach writer on 
the subject, on getting from a Highlander the 
correct Gaelic form of a certain combination 
which he meant for the explanation (by appear- 
ance) of a certain place-name, rejected this 
correct form of spoiling his derivation, and kept 
his own original wrong combination ! Such a 
scientist as the late James Macdonald of Huntly, 
who honestly tried to acquire the language, 
never attained complete correctness in repro- 
ducing Gaelic names pronounced to him by the 
natives. What with the Gaelic article causing 
aspiration and eclipsis, bewildering to a non- 
Celt, and the other phonetic and syntactic finesse 
of a language which has undergone more than 
ordinary philologic change, Gaelic is a lan- 
guage which only a very well trained outsider 
can have anything to do with. This training 
our place-name philogists as yet refuse to under- 
go. And there are also the history of the Ian- 


guage — its changes through hundreds of years 
— and the history of the country during the 
same time, all to be taken into account. It may 
truly be said that the writer who undertakes to 
deal with the Celtic place-names of Scotland 
must undergo no ordinary linguistic and historic 

Mr Watson fulfils all the requirements of the 
philologist we need to elucidate our Celtic place- 
names. His Introduction of some hundred 
pages is a mine of practical information, thor- 
oughly systematised. In dealing with Gaelic 
names, the student will be first struck with th^ 
large place which he gives to suffixes To the 
ordinary philologist every ending in ach is for 
achadh, " field "; Mr Watson shows this suffix 
to be old Celtic -cicum, denoting " pla^.e of,'* 
such as Carn-ach, " place of cairns," or Dorn- 
och, " place of hand-stones." He shows with 
clearness how suffixes combine : Muc-ar-n-aich, 
" place of pigs," where we have three suffixes 
{ar, an, ach). One important point which he 
brings out is the undoubted existence of a 
diminutive -aidh or-idh, at least in old Pictland. 
He adduces lochaidh, badaidh, and lagaidh as 
outstanding examples. These suffixes seem to 
be the old Celtic ending in -io-s, or fern, -ia, the 
latter very common in river names. The diffi- 
culty here, however, is the modern Gaelic pro- 



nunciation in final -idh, not -e as in usual GaeJi<^ 
The Welsh, however, pronounce, or rather 
spell, this ending {-io-s) in its modera form ns 
-ydd. It would seem that in this diiuinutive 
ending -aidh we have distinctive, traces of Pictish 
or Brittonic pronunciation of these place-names. 
We have such diminutives m old Gaelic in cer- 
tain personal names, such as -Ba-^i e (St Barr) for 
Barrio-s, and this again for Barro-vindos or 
Barrfhind, which we know to be the fuli name 
of the Saint. Those acquainted with the old 
charter forms of place-names know that -ie, the 
Scottish form of G. -aidh, is continually inter- 
changed with -in. This last, which does not 
usually exist in a Gaelic form, must be the old 
Pictish stem-ending (from -id, gen. -inos) in n, 
known well in Scottish Gaelic, and giving rise 
to the modern Gaelic plural, just like the weak 
stems in the Teutonic languages. Material for 
pursuing this and kindred points will be found 
in abundance in Mr Watson's volume. We 
may also note his excellent tabulation of Norse 
vowels and consonants in Gaelic; it should be 
very useful to students of Northern names. But 
does not Homer nod in explaining Saraig as 
Saur-Vik (Mud-bay), the phonetics of which by 
the table result in Soraig? Compare Soroba^ 
Sorby, and English Sowerby. 

The main body of the work deals seriatim. 
with the twenty-nine mainland parishes of Ross 


and with Lewis in general. Each parish forms, 
as it were, a chapter by itself; the place-names 
are dealt with in separate articles, vocabulary- 
wise, but not in alphabetical order. The 
' ' English ' ' or map-name is given first ; then the 
old forms from charters, documents, or his- 
tories; then the Gaelic form, where such is 
existent; and, lastly, the meaning or derivation. 
Mr Watson has heard all the pronunciations per- 
sonally, and he has visited practically every 
corner of the county. On this head th'-3 work 
is most thoroughly done, and the derivation 
offered suits the characteristics of the place, if 
it be named after any characteristics. The 
county name Boss he is inclined to derive from 
Brittonic or Pictish sources, corresponding to 
Welsh rhos, " a wold," rather than from Gaelic 
ros, " a promontory," but the words are no 
doubt ultimately the same. Cromarty contains 
the adjective crom, " bent," but the old forms 
are puzzling, and the modern Gaelic Cromba' 
points only to crom-hath, "curved sea." Mr 
Watson restores the old name as Crom-b-ach- 
dan, the h being a development (of Pictish 
times?) and the rest mere suffixes, the total 
meaning " Bay Place." Pictish, Norse, and 
Gaelic names jog one another all over the 
county, but, as the author well shows, there is 
a marked difference between Easter Eoss names 
and those of Wester Eoss, the latter being more 


Gaelic and more modern really. The Norse 
element stops at the Beauly and Tarradale and 
Eskadale (Ash-dale, lately explained as Uisge- 
daill). Pictish names are common in East 
Eoss. One of these we have in Bal-keith, 
doubtless for older Pit-keith; Gaehc, Baile-na- 
Coille, a translation which, as Mr Watson points 
out, seems to prove that Keith means " wood," 
from a word aUied to Welsh coed, "wood"; 
Gauhsh ceto-, allied to Enghsh heath. Dal- 
keith is therefore Brittonic in both elements, 
" Plateau of the wood " (Welsh dol, Pictish dul, 
dal). Space does not allow us to follow Mr 
Watson further in quoting his interesting deriva- 
tions, but we must mention some old or peculiar 
words which he has been enabled to recognise 
or rescue. Strikingly happy is his derivation of 
the place-names Nonakil (" church-land "), 
Newmore, Dalnavie, and Navity, from the old 
Gaelic nemed {neimhidh now), " a sacred 
place," which we have also in Rosneath. 
Eirhhe or airbhe, '" a wall," is found in Altna- 
harrie, etc. ; rahhmi, a kind of bulrush; saothair, 
a neck that joins a " dry-island " to the shore, 
a promontory covered at high tide; faithir, the 
steep face of an old raised beach; feodhail, a 
side form of faodhail, " a ford," from Norse 
vadhilL ''shallow water"; sleaghach, a rifted 
or gullied slope or hill, from the same root as 
sl'ujhe, "path," literally "a cutting" (root 


sleg, "hit," "cut"). On the west coast 
cathair means a " fahy knoll," while sithean 
means a considerable hill with no notions of 
fairies attached thereto. Mr Watson rightly 
queries druineach as Druid; the meaning is 
artist, artificer, sculptor (Mr Carmichael's 
draoineach). Irish druine means " art," even 
"needlework." We demur to Mr Watson's 
derivation of Killearnan. True the Gaelic is 
Cill-Iurnain, which might point to a St Iturnan, 
only the name Iturnan is a misreading for 
Itarnan, a true Pictish name and a saint's name 
also. Ernin or Ferreolus was a favourite saint 
and a favourite saint's name, and to a root- 
inflected form of Ernin or larnan we must refer 

Mr Watson has added a valuable index to his 
work, containing over three thousand words, 
and showing by a device with the full stop where 
the main accent rests. Many of these words 
naturally Ijelong to districts outside Eoss. In- 
deed the volume, as already said, is a microcosm 
of Scottish place-names, the Anghc Lothians and 
the Merse being left out of account. It lays a 
sound basis for the further study of Scottish 
place-nafnes on modern philologic lines. 




The Place Names of Elginshire. By D. 
Matheson, F.E.I.S. Stirling : Eneas Mac- 
kay. 1905. 

Mr D. Matheson, lately head of an educational 
institution in Elgin, and now editor of the 
Northern Times, is the latest recruit to the slowly 
increasing authors on place names. His work 
on the " Place Names of Elginshire " is a hand- 
some volume of over two hundred pages, pub- 
lished by Eneas Mackay, Stirling, and dedicated 
to Mr Carnegie. Works on place names have 
steadily been getting more scientific as their 
authors have studied the science of language and 
the possibilities which the history, the physical 
features, and the languages of the localities dealt 
with afford. The authors have also profited by 
reading what has recently been done by others 
in the same line. Mr Matheson, however, 
stands by himself, and is a law to himself on 
language and history. He has consulted Skeat's 
Etymological Dictionary, but not with profit; he 
does not know that there is a corresponding 
Gaelic Etymological Dictionary, considered of 


equal authority according to leading Celtic 
scholars. The latter 

' ' Wad from many a blunder free him 
And foolish notion," 

No doubt Mr Watson's book on the " Place 
Names of Ross-shire " came out too late to be 
of use to him; this is a pity, if Mr Matheson 
would have utilised the methods and results 
attained there — which is doubtful. Mr John- 
ston, of Falkirk, he does quote; it is a pity that 
he did not use even this indifferent work to 
better purpose. In short, Mr Matheson belongs 
to the old school of etymologists. He has no 
compunction to refer a Moray name to a Ger- 
man, Dutch, or Scandinavian origin straight 
away, without considering the question of how 
the Germans, for example, could ever have 
planted a name in Moray. Thus, for Knock- 
granish, near Aviemore, he gives: "from the 
Gaelic * Cnoc,' a hill, and the Teutonic ' Gran ' 
or ' Grense,' a boundary," the latter being a 
German word. How could he get his Germans 
up to the heights of Craigellachie several cen- 
turies ago? Besides, Granish is famous in works 
on Druid lore, as our author should have known. 
Again, Balvatton, in Cromdale, is taken from 
Gaelic ' baile ' and Norse ' vatn,' water. Were 
the Vikings anytime about Grantown? Of 
course, the name means ' Town of the clump 


(badan).' In his Scandinavian eagerness, he 
antedates them by 600 years. Lossie, Ptolemy's 
Loxa, dates to 120 a.d. at least; the Vikmgs 
came in 800 a.d.; yet our author calmly asserts 
that Loxa is Laxa of Norse, ' salmon river ' ! It 
means ' winding river ' (Old Gaelic lose). In 
matters Gaelic, Mr Matheson belongs to the 
school of the late Colonel Eobertson, famed for 
his " Topography of Scotland." We thought 
that Mr Matheson knew Gaehc; we are now 

In history Mr Matheson's work is " second 
hand of second hand," and his introduction is an 
extraordinary jumble. Lollius Urbicus, about 
whom only a few words exist recording the fact 
of his building the thirty-two mile wall, is repre- 
sented as conquering to the Beauly Firth, calling 
the district southward Vespasiana ! Where in 
the world did Mr Matheson get this utteiiy 
absurd statement? His confusion in regard to 
Ptolemy is quite inexcusable, as excellent edi- 
tions of that author's Geography of Scotland can 
be easily got at (Gaelic Society's Trans. XVIII. ; 
Proc. Society of Antiquaries, XL, by Captain 
Thomas). There he could see that Burghead 
was in Ptolemy's original ' Greek ' (not Latin) 
Pteroton Stratopedon, or ' winged camp,' and 
that such ' fool ' work as Tor-an-duin, founded 
on Ptolemy's word Pteroton, was beneath notice. 
Mr Matheson has the Scandinavians on the brain. 


One third of his Moray names, speaking roughly, 
he refers to Scandinavian sources ; now it is safe 
to say that not one name in Moray is due directly 
to the Scandinavians. They never colonised 
there, and the only recorded battle fought by 
them in Moray was that of Torfnes, in which 
they defeated King Duncan (probably). The 
assertions of the Saga, when examined judicially, 
show that the Norse made a ' ' blood-red ' ' foray 
through Moray to Fife. Undoubtedly they held 
part of the old province of Moray — the Ross 
portion — and this is the foundation for the Saga 
statements. From place names we know they 
stopped at Beauly river, the old limit of the pro- 
vince of Ross. All Mr Matheson's Norse deri- 
vations may be unmercifully excised, but with 
what a result to the book ! That there are Norse 
and Dutch names in Moray we do not doubt ; but 
these were brought in by the English and Nor- 
man-French planted in the district in the 12th 
century to replace the transported natives. This 
is historical, and if Mr Matheson had studied the 
Teutonic names in Moray with these facts of the 
12th century in his mind, he would certainly 
have added to our know^ledge both of place 
names and history. A few of the names brought 
in by these colonists from Lothian, Danish Cum- 
berland, and, even, Flanders, may be pointed 
out. Hatton is in Moray three times — ' heath- 
ton ' (Taylor) ; common in England ; Overton, 


Harestanes, Middleton, Whiterigs, Unthank 
(useless land), Oakenhead or Aikenhead (Blaeu), 
Orton, Ogston, Coleburn (compare Cold-beck of 
Cumberland), Ormiston, Mundale (Mundwell or 
' Inlet-field ' over the borders, whence the Bor- 
der surname Mundel — Norse in origin), and 
several others. This would have been a fertile 
field of research; but Mr Matheson's Scandi- 
navian wet blanket is over it all. 

Mr Matheson wastes much space by giving 
the derivation not only of the place name, but 
the derivation also of the constituent words. 
Thug, if a place is called ' Hill-head,' surely for 
ordinary purposes it is sufficient to say " from 
Eng. ' hill ' and ' head.' " No ! Mr Matheson 
must etymologise both ' hill ' and ' head ' — 
thus : — 

" HiUhead. — Middle English hil, hul, Anglo- 
Saxon hylJ, Dutch hil, Latin collis, Lituanian 
[sic.'] Kalnas, a hill, and head from the Middle 
English hed, heed, heued, Anglo-Saxon heafod, 
Dutch hoofd, Icelandic hofud, Latin caput, 
Greek Kephale [sic' really alhed toEng. <ja6/e], 
Sanskrit Kapala, literally a skull, and by usage 
n head, or end." 

Such is the passage, with all its imperfections of 
typing (not due to the printer) and derivation. 
If Mr Matheson had dropped these useless addi- 
tions to simple hill and head, he would have re- 
duced his book by about one half. But the 


worst of it is that he iterates these Usts every 
time the respective words turn up, and if the 
derivations are all awry, as they are usually in 
Gaelic, the constant repetition is doubly irritat- 
ing. The Gaelic word lag, with its derivatives 
lagan and logie, occui' very often, but nearly 
every time we have this piece of atrocity 
attached : — 

" Gaelic lag, lug, German lucke, and cognate 
with the Latin Lacus, and Greek Lakkos, a 
hollow or lake." 

Now it is just possible that ' lag ' may be allied 
to German ' lucke,' gap (Kluge is doubtful about 
its derivation), but certainly it is not allied to 
either of the Latin or Greek words, which agree 
with Gaelic ' loch.' Mr Matheson has been very 
unwise to touch Gaelic derivation, considering 
his knowledge of the subject, and the further 
fact that his particular derivation work is already 
competently done and placed before the public 
in dictionary form. 

Some other leading errors must be pointed 
out. The termination '-as,' '-us,' '-ais' is 
common in Moray, as it is in all Pictland. Mr 
Matheson, in an evil moment following old Shaw, 
made this to be ' eas,' waterfall, though Shaw 
adds the idea of ' water ' generally. The num- 
ber of waterfalls in Moray would have been 
counted by scores were this derivation true. 
The sufRx denotes 'place,' ' station,' and is no 


doubt derived from the root ' ves,' ' vos,' dwell, 
be, allied to Greek ' astu,' a city, Sanskrit 
' vastu,' a place, and perhaps to Gaelic ' fois.' 
Dallas means ' plane place,' from old ' dul,' now 
' dail ' ; Eothes, ' place of raths or granges ' ; 
Duffus, 'dark place'; Forres, 'lower place,' 
Pictish ' voter,' ' fother,' ' for,' ' far ' ; Forres, 
Mr Matheson says, appears in Ptolemy as Varris 
— which is not the case ; Granish, ' rough place ' 
(gran), not ' grense ' of German (Mr M.), nor 
' grian' of the Druidists; and here add " gourish,' 
of Ach-gourish, etc., ' place of goats.' Simi- 
larly Pityoulish is for Pit-geldais, the root ' geld ' 
(Geldie, etc.) denoting ' water,' as in old Eng. 
' child ' (Chillam), a spring, Norse ' kelda,' 
spring, whence St Kilda. The prefix ' lyne ' 
appears very often in the Strathspey portion of 
Moray. Mr Matheson derives this from ' linne,' 
pool. It is safe to say, but with one exception, 
not one of the many names so prefixed comes 
from 'linne.' The word is 'loinn,' locative of 
' lann,' a land, a glade, Welsh 'llan,' of which 
our author speaks so often. Here Mr Matheson 
has failed in his duty towards Gaelic and local 
pronunciation : ' loinn ' is very different in pro- 
nunciation from ' linne,' and the mistake is inex- 
cusable. Besides, there are no pools near most 
of the places so named; here Mr Matheson fails 
in his facts. As a matter of fact, he does not 
know the County from a linguistic standpoint. 


His Gaelic derivations are naturally best 
tested by the Highland border parishes— Aber- 
nehy, Duthil, Cromdale, and Edinkillie. The 
' abpT ' of Abernethy is from ' od-' or ' ad-ber,' 
Lat. ' af-fer,' root 'fer,' 'ber,' bear; "there 
never has been any doubt " about this among 
m«)de¥n philologists till Mr Matheson has cast it. 
Aiiernethy is derived by Mr Matheson from Aber 
n-aitionn, ' confluence of the broom ' ; we can- 
not characterise this piece of ineptitude. The 
rivers Lochy and Lochty come from Loch-dae^ 
' Black ' (loch) and ' dae ' (goddess), not from 
Lochdubh, as Mr Matheson says; for Adamnan 
distinctly speaks of ' Nigra dea ' as the transla- 
tion oi the Lochaber Lochy. ' Tobar/ well, is 
not the same as ' dobhar,' water, nor as ' tiobar,' 
well, though the first and last are both from the 
root * bhru,' spring. Auchtercheper, in Duthil, 
is for Achadh-da-tiobar, 'field of the two wells,' 
not My M.'s *uachdar-ceap-tir,' which is impos- 
sible in view of the Gaelic sounds. Ry-voan, m 
Abernethy, is ' Bothie-reach,' not 'peat-reach.' 
Mr Matheson spells this common word Ry very 
badly as ' reidh ' ; it is ' ruigh,' a stretch or piece 
of land at the base of a hill, also ' a fore arm.' 
Causor, in Abernethy, is Cabhsair (causeway), 
not ' Casair.' The Desher of Duthil is not from 
Lat. * disertum,' but from Gaelic ' deisear,' 
south side, Loch-tay Disher, where also is old 
Toyer, still 'tuathair.' Seemingly Duthil is an 


assimilation of Tuathail (north side), to Desher 
on the south side of the same hill. Carr-Bridge, 
Drochaid Charra, is from ' cartha,' a pillar stone, 
old Irish ' coirthe.' The pillar stone or stones 
are there. Slochd surely is from Gaelic ' sloe,' 
a pit, gully. Gallovie, as in Laggan, is a deri- 
vative of ' geal/ ' White-land.' Lochneiian, 
Loch-an-ellan, is ' Loch with Isle.' It is curi- 
ous what a penchant the amateur philologist has 
for ' ailean,' green spot, which is really rare in 
place names. So is ' aite,' place; it scarcely 
occurs, yet according to the amateur it is every- 
where. Inverallan is from Allan river; this 
river-name is common in Pictland and Welsh- 
land, and is possibly from the root ' pal,' Lat. 
' palus,' marsh. Anyway, it is not from ' ailean,' 
a green. The derivation given for Kriockando 
— ' Cnocan-dubh ' — shows that Mr Matheson 
never heard the Gaelic of the name, which is 
Cnoc-cheannachd, 'Market knoll.' Even the 
Sassenach etymologists know this, and use it as 
a warning name ! Dalchapple is derived by Mr 
M. from ' caibeal,' a chapel, which in Scotch 
Gaelic is ' seipeal.' Of course the name means 
here and elsewhere ' Dell of the horses (capull).' 
Fionnlarig is ' White pass,' from ' laraig,' a 
pass, ' learg,' hill-side. 

Mr Matheson, of course, gives himself un- 
bridled license in regard to hybrids; Gaelic, 



English, and other Teutonic roots are welded 
together indifferently. The worst case is that 
of Ptolemy's Tvesis, now known to be Spey 
(Scotch ' spate ' allied). This Mr Matheson 
writes Teussis, and derives from the Greek 
Teukrion; on what principles, we wot not. 
Where a word begins as Gaelic, Mr Matheson 
should remember, it will end in Gaelic. A 
funny hybrid is recorded in Pit-airlie; this is 
Pictish ' pet ' and Eng. ' early ' ! He shows a 
preference for the out-of-the-way rather than 
the evident etymology. Cummingston is from 
St Cummein, not from the Cummings. Is this 
likely? Blinkbonny appears thrice, each time 
with a nevr derivation; first, French 'blanc,' 
white, Gaelic 'ban,' hill [sic]; second, Fr. 
'blanc,' G. 'ban,' white ('White-white'!); 
third, as ' blink ' of Eng. and ' bonny ' of Sc, 
which, of course, it really is. Rev. Mr John- 
ston gives several other places so named, and 
translates it ' Belle-vue.' The Pictish ' pett ' 
or ' pit ' is not allied to English ' pit ' (p. 150), 
as the modern discussions on the Pictish ques- 
tion should have taught Mr Matheson. The 
places called Bauds seem to be from the plural 
of G. ' bad,' a clump — ' place of clumps.' Very 
many other mistakes could be pointed out in the 
names of the Laigh of Moray, apart from the 
Highland parishes. Elgin, as a name, cannot 
be separated from Glen-elg ; the word ' elg ' in 


ancient Gaelic means ' noble.' As for Moray, 
its oldest forms are Mureb, Norse Masrh^fi, now 
Morro or Mortho. The whole points to a Pictisli 
Moriti, dative pi. Moritobis, whence Murref of 
early documents and Norse Magrh^eli. It would 
aspirate into Morthaibh, ace. Morthu (Pictish 
Moritos), admirably suiting the modern phon- 
etics. The root is ' mor ' of ' muir,' sea, and 
the (jraulish tribal name Morini is its exact 
parellel. The meaning is ' Sea-side folk.' The 
exact value of Mr Matheson's book we care not 
to assess. Of course, every one interested in 
place names should have it. If he had given 
the old forms of the names with dates, this in 
itself would be of great value ; if he had indi- 
cated the pronunciation with accented syllable, 
this also would be, pace Mr Matheson, of great 
importance. But his references to old forms 
are vague and sometimes misleading, as when 
he means by old or primary form what he con- 
ceives to have been the old form. As a work- 
ing list of names, with some historic facts, the 
book will do. 




A Review of " Place-Names of Scotland "• 
by the Rev. James B. Johnston, B.D., 
Falkirk. Pubhshed by Mr David Douglas, 
Edinburgh, 1884. 

*' Place-Names of Scotland " is the name of 
a book by the Rev. James B. Johnston, B.D., 
Falkirk, publislied by Mr David Douglas, Edin- 
burgh. The proper title of the book should 
have been something like that of Colonel 
Robertson's work on the Gaelic Topography of 
Scotland. Both books deal mainly with the 
Gaelic names of Scotland, real or supposed, and 
both are equal in philologic value. Indeed, Mr 
Johnston's work is distinctly worse than his pre- 
decessor's on many points, despite his having 
the advantage of several works that have 
appeared since Colonel Robertson's pioneer 
book of over twenty years ago. Mr Johnston 
has at least a nodding acquaintance with Pro- 
fessors Mackinnon and Rhys, and, really, from 
the state of philologic knowledge in Scotland at 
present, we should have expected much better 
results. He confesses to only an amateur's 


knowledge of- Gaelic, but he might have saved 
himself the trouble of the confession. His work 
too }3lainly reveals the fact. And w'e need not 
wonder at his saying in regard to Gaelic spelling, 
" that there is probably no language in the 
world in which the eye can give less help to the 
tongue." This is simply nonsense. Few lan- 
guages are written more strictly according to 
rule, and, if Mr Johnston had only taken the 
trouble to master these rules, he would have 
saved himself from any amount of bad etymolo- 

Gaelic is only one portion — though the 
largest — of the equipment necessary for one 
that is to tackle the place-names of Scotland. 
A knowledge of the principles and practice of 
philology is absolutely necessary; and, though 
Mr Johnston knows some philologists like Pro- 
fessors Ehys and Mackinnon, and quotes them, 
yet personally he knows nothing of philology. 
When he has to trust to his own unaided re- 
sources, the result is philologically lamentable. 
Again, a clear conception of the history of 
Scotland is necessary for the place-name ety- 
mologist. Were the Gaels really the first 
inhabitants of Scotland? and is one justified in 
looking for Gaelic place-names in South-Eastern 
Scotland? Is it not the case that Gaelic did not 
penetrate south of the Forth until the Macalpine 
dynasty (844-1033)? We omit Galloway, and 


the western coast as far as the Clyde. The 
Gaehc place-names that occur in the Edinburgh 
district clearly belong to Gaelic that had already 
borrowed deeply of the Norse. The prefix dal, 
a dale, is a Gaelic word borrowed from the 
Norse; and the Dalkeiths and Dalrys sOuth of 
the Forth prove that they were so named much 
later than the year 800, when the Norsemen 
came first. Another point is the Pictish ques- 
tion. Were the Picts also Gaels? Skene, of 
course, says they were : they spoke a low Gaelic 
dialect, he says — whatever that may mean. But 
Mr Johnston, who accepts Skene's views, does 
him a real injustice by supposing that he ever 
maintained the absurd idea that the Cornish was 
a Gaelic language. That language is Brythomc 
of the Brythonic. Could Mr Johnston not 
decide that point himself by testing the language 
philologically, instead of pitting Skene against 

Further, Mr Johnston has not made himself 
acquainted with all the literature of his subject. 
Professor Mackinnon's articles on Argyllshire 
Place-Names he knows, and also Sir Herbert 
Maxwell's Topography of Galloway. It is 
doubtful if he has consulted Captain Thomas's 
two contributions on the Place-Names of the 
Hebrides and of Islay, in the Society of Anti- 
quaries' Transactions. They are, outside Pro- 


fessor Mackinnon's articles — and he owes much 
to Captain Thomas — the best thing done in 
Scottish topography. Mr Johnston has over- 
looked many of Captain Thomas's derivations, 
which are correct, while his own are not. 
Further, many papers have appeared m the 
Transactions of the Inverness Gaelic Society on 
this subject, all of them being good, some being 
excellent. Dr Cameron's paper on Arran 
Place-Names appeared in Volume XV. It would 
have saved Mr Johnston from etymologising 
Lamlash as Lan-Maol-Iosa, for it was the island 
that was called Lamlash, that is, Eilean-Molais, 
the Isle of Molas, a well-known Irish saint. We 
may mention Mr Mackay's excellent series of 
papers on Sutherland Place-Names in the same 
Transactions, Mr Macbain's Badenoch Place- 
Names, and Mr Maclean's papers on Alness and 
Kiltearn. Besides, Dr Whitley Stokes passed 
in review the old Pictish names in the Philo- 
logical Society's Transactions last year. He 
clearly showed that aher and pet or pit, as pre- 
fixes, are Pictish, and that Pictish belongs to 
the Welsh group. Besides, Pictland extended 
from the Forth to the Orkneys, and Pets and 
Abers can be traced as far north as Sutherland 
and as far west as Drumalban. 

Mr Johnston opens with some chapters on 
general principles, where he passes in review 
the characteristic of Celtic, Gaelic, Norse, and 


Other place designations. We need not speak \ 
of his philology. Suffice it to say that he equates i 
the Welsh pen with Gaelic hen, and concludes, 
like Skene, that Welsh p may. appear as Gaelic 
b. Now, Gaelic heann has a very good Welsh 
equivalent in Welsh ban, of like meaning and 
use; while pen is the corresponding form to 
Gaelic ceann or Kin-. Gaelic final r is never 
lost like the Norse r ; so Geldie cannot be Gelder, 
nor Orciiy be iirchar, a cast. Aher is not 
strengthened either to ar or ah, but a syllable is 
left out, if it repeats the same sound as the 
previous one, or there is a metathesis of the r 
as in Arbroath. Hence the Aberbreachy of 
1334 loses one of its hers and now appears as 
Ahriachan, which Mr Johnston absurdly etymo- 
logises into Ahhriabhach, grey watei . And 
here we say something of Mr Johnston's abh and 
an for water. They are mere figments of his 
oion and other etymologists' imagination. No 
such words exist or have existed in the Gaelic 
language, and yet several scores of Mr John- 
ston's derivations depend upon these words ! 
The an which terminates several river names is 
a mere adjective termination seen to advantage 
in the Gaulish river-names Sequana or Matrona. 
The word aoi, isthmus, is a Norse word, and 
cannot form the root of lona, nor can i, island, 
Wiiich again is the ey of Norse borrowed. Mr 
Johnston's lack of knowledge of Gaelic makes 



him use the absurdest of GaeHc words, so long 
as the sound suits. Some of the words he offers 
are obsolete, some rare, and some are, as 
already said, mere figments of the dictionary- 
maker. The Gaelic uisg, water, has nothing to 
do with the Pictish Esk, Ptolemy's Iska. Mr 
Johnston's theory about Hybrids is ridiculous, 
and both his examples are demonstrably wrong. 
Newtonmore, where he regards more as Gaelic 
mor, big, is really Newton-muir, the Muir's 
Newton — Baile-iir-an-t-sleibh in Gaelic ! His 
Garrabost is explained by Captain Thomas as 
Geirabost or Geirr's Bost or Farm, and not 
Garbh Bost, the rough farm. If Mr Johnston 
wants a proper hybrid he must look for forms 
where the sense of the older word has been 
forgotten. Such a case occurs in a name like 
Strath-halladale, where the idea of dale is re- 
peated twice, the Gaels not recognising that the 
terminal dal meant a dale. 

We may now take a few specimens of Mr 
Johnston's derivations, beginning at Inverness. 
He suggests that Ness may be from nios, from 
below; but nios has the long i sound and the i 
of the Gaelic for Ness is short. We need not 
speak here or elsewhere of how inapplicable the 
sense of the Gaelic words offered is to the river 
or place-name. Clachnacuddin is etymologised 
as Clachcudachan or St Cuthbert's stone, not 
Clach nan cidainn, the tubs' stone. Clachna- 


harry fares equally badly : the root is given as 
carraid, strife, instead or aire, watching. Nairn 
is for an am, the flank, where the word am is a 
figment of the lexicographer. Ross means really 
a wood or moor, not a promontory. Moray is 
of course Mor ahh, big water! Yet it is clear 
that the root is mor, sea, for the vowel is short. 
Aberdeen may represent the confluence of either 
Dee or Don, it seems. The Dee he takes 
from the Gaelic Deahhadh, draining, though 
Ptolemy's Deva makes it clear that it and his 
Devana or Divona, now Don, both mean " god- 
dess," and indicate the existence of river 
worship, of which we hear from Gildas. 

' ' Philology founded on sound is not sound 
philology," and when Mr Johnston etymologises 
Knockando as Cnocan-dubh, black hillock, he 
commits two blunders : he should attend to the 
place of the accent on words ; that shows where 
the main root of the word lies, the accent being 
in this case on the and; and he should know that 
the real Gaelic name of the place is Cnoc- 
cheannachd or Market hill, as old Shaw carefully 
explained over six score years ago. The Gaelic 
accent is of course on the first part of Ceann- 
achd. Aviemore is abh mor, big water; but the 
Gaelic pronunciation is Agaidh mhor, the big 
agie, whatever that may be. Balintore is given 
as Baile an Deoraidh, the Dewar's town, but the 
name really means the town of the Bleaching. 


Balfour, and the other words in four, is a glar- 
ing example of bad etymologising. In tiiese 
words, the accent is on the four, which clearly 
shows that this is the main root. But Mr John- 
ston explains four as for fuar, cold. Yet Dal- 
four, if for cold dale, would undoubtedly m 
Gaelic be Dailfhuar, that is Dal-uar, with the 
accent on the Dal and the / entirely gone. The 
four of these words must stand for pour, a 
Pictish word denoting pasture land — if we may 
guess from the Breton peur, Welsh paivr. This 
etymology, suggested in " Badenoch Place- 
Names," has been accepted by Dr Whitley 
Stokes in his revised edition of his Pictish voca- 
bulary. Curiously the prefix both, habitation, 
which is so common in Gaelic Place-Names, 
finds scarcely a place in our author's work. Yet 
Boleskine, Balquidder, and several others show 
this prefix. 

Mr Johnston is not satisfied with one or tvro 
root words, in explaining a name of any length, 
and he generally manages to stick on an extra 
abh, an, or ach (for achacUi, held). Cabrach is 
for cabar-achadh, deer field, whereas the ach 
here and elsewhere is the adjective termination. 
Conan is for con-an, or caoin abhuinn, gentle 
river, which, as a matter of fact, it is not. 
(dencoe is variously derived from cu, cow, dog, 
or from coill, wood, or from cornair, confluence. 
Yet the Gaelic name is Gleann-comhann, or 


narrow glen. Mr Johnston's alternative deri- 
vations are always irritating, and the fact of his 
giving them shows how utterly unreliable and 
unscientific his work is, Culloden is given as 
Cul-lodan, back of the pool; but the Gaelic is 
Cuilodair. Culross is anciently Culenross, or 
Holly- wood. Dalarossie is not the field of the 
ros or promontory, but Dail-fhearghuis, or 
Fergus' dale. Dalnavert is not Dal-na-bhaird, 
Bard's dale, but Dail-nan-feart, dale of the 
graves. The word dahhach, four ploughgates, 
is not from damh-ach (sic !), ox field, but from 
dahhach, a tub or corn measure. Dulnan is not 
Dail an an, river dale, but Tuilnean, from tuil, 
flood. The Earn is not Ear an, east flowing, 
for the Gaelic is Eire, with long e, and its geni- 
tive is Eireann, the same name as Ireland. 
Auldearn is in Gaelic Allt-Eire or Earn Stream. 
Feshie cannot be from fdsach, desert; it is a 
river name, and the root vowel is short e, not 
long a. Garry cannot be from garhh, rough; 
the Gaelic is Garadh. The Gaelic garradh is 
merely the English garden borrowed. Gask is 
not for crosg, a crossing. Urquhart is absurdly 
explained by Ard-a'-cheaird, smith's height. 
Nov;, Adamnan, about 700, gives this name as 
Airchartdan, where the air is clearly the pre- 
position, and the root word is cartd or card. 
With this name we must connect Kincardine, 
which seems a half Pictish word for ' ' end of the 


rowan wood"; not cinn gairdein, head of the 
arm, as Mr Johnston has it. We might deal 
thus with over a third of Mr Johnston's some 
three thousand words. The work is a mass of 
guess-work, slavishly following the spelling, and 
forgetting too often the history of the word or its 
present sound. Indeed, without indicating the 
modern sound of the word and the place of the 
accent, such books as this are worthless. Mr 
Johnston ^ives from the Origines Parochiales, 
the oldest forms of many names, and we must 
say that this is the most valuable contribution 
he has made to the elucidation of the Place- 
Names of Scotland. 





The Place-Names of Argyll. By H. C. 
Gillies, M.D. London: D. Nutt. 1906. 

Dr Gillies' work lias been received with a 
universal chorus of praise by the press, from the 
London " Tribune " to the Ohayi Times. The 
reviews were commendably short, for they 
showed no marks of familiarity with Gaelic 
place-names. There has been, therefore, no 
expert opinion offered as yet to our knowledge; 
and as Ian Maclaren has just said, ' ' the present 
day is a day of experts; the day of amateurs is 
past," adding that in any subject we seek expert 
advice when we wish to know. There are very 
few experts in Celtic scholarship or in Gaelic 
scholarship to-day in Scotland, but their number 
is increasing with fair rapidity, thanks to our 
Celtic Chair. In this work on Argyll Place- 
Names, Dr Gillies comes forward on his own 
credentials as an expert in Gaelic philology in 
its most difficult aspect, that of elucidating place- 
names. He says : — " I am quite aware that the 
work is far from perfect. No person could make 
it perfect; and certainly no one in my position, 
with my poor scraps of available time, could do 
it better. I believe it is as nearly correct as any 


one could make it." Now here Dr Gillies gets a 
little mixed in his climax. First, he can do it as 
well as any man in his position as regards time , 
second, the work is as good as any man can do, 
time or not! Really the Doctor protests too 
much. The book, however, takes the same Sir 
Oracle tone throughout. The work of previous 
writers he ignores or overlooks, except in one 
or two cases. For instance, a careful study of 
Mr Watson's work on the Place-Names of Ross- 
shire, which, by its excellent introduction is at 
once a text book and an example book for the 
study of Scoto-Celtic place-names, would have 
saved Dr Gillies many absurdities in his Norse 
etymologies; for Dr Gillies has quite a craze for 
explaining names as of Norse origin. A feature 
of the work is ils perversity; long established 
etymologies are thrust aside for something 
new or bizarre (as in the case of the 
county name), or the obvious derivation is 
overlooked, generally for a Norse one. Again, 
the format of the book is bad. What is 
wanted is to give first the map or post-official 
name; then the modern Gaelic pronunciation; 
thirdly (if the word is difficult) its oldest forms 
and changes; lastly, its derivation, with proof 
adduced, such as the suitability of the explana- 
tion to the character of the place. Then the 
etymologist should, if possible, see the place and 
hear the name pronounced, or at least he should 


get a description of the place and hear the pro- 
nunciation. As regards the ancient forms, Dr 
Gillies takes up the monstrous position that they 
are not necessary, and that, too, when he is 
professedly dealing with Norse words and diffi- 
cult Gaelic words. He certainly saves himself 
much research, but at what risk? Who would 
for a moment think that modern Askary stands 
for Norse Asgrims-erg, the " airigh" of Asgrim? 
Or Scrabster for Skara-bolstadhr ? Then again 
Dr Gillies has clearly trusted for his form of the 
name to the ordnance map m too many cases; 
this is evident, and he admits it in some cases. 
The motto with the expert in place-names is to 
accept no ordnance map name unless it is veri- 
fied. Dr Gillies has had some predecessors m 
the field of Argyll Place-names. Piofessor Mac- 
kinnon wrote a series of eighteen articles in the 
Scotsman on the "Place and Personal Names 
of Argyle " in 1887-8, when he showed 
the sound, sane scholarship and literary exposi- 
tive power that ever since has marked his work. 
And in place-names, sanity and scholarship must 
conjoin ; running after the bizarre or fanciful is 
fatal. Rev. Mr MacNeil's "Guide to Islay " 
contains mostly expert-produced derivations, the 
late Hector Maclean, Captain Thomas, and Dr 
Macbain having helped. Drs Reeves and Skene 
went over the names of lona and Tiree, making 
valuable lists and exhuming old church names. 


So much for preface. I he name Argyll is in 
G., Earraghaidheal ; early Gaelic, Airer-gaidhel 
(Annals of Ulster), i n-airiur Gaoidheol (Three 
Fragments), etc. The word here is " oirear," 
district, coastland; Irish, oirear, early Irish, 
airer. M'Vurich speaks of Argyll as being 
divided into two districts — " Oirer a deas," 
Argyll proper; " oirer a tuath," North Argyll to 
Lochbroom. The name means the " Coastland 
of the Gael," or, as the 12th century writer puts 
it in Latin, " Margo Scottorum." The Gael in 
Ireland and in Scotland bore one or two names, 
Scot and Gaidheal being the favourites among 
the people themselves. The present view of 
Celtic scholars accords with the old annals, not 
with Skene and later writers. Professor Kuno 
Meyer writes " that no Gael ever set foot on 
British soil save from a vessel that first put out 
from Ireland." The annals put the first invasion 
of the Scots about 160 a.d., the leader being 
Cairbre Eiada, son of Conaire II., King of Ire- 
land. Other invasions followed, and the Scots 
and Picts joined to attack Roman Britain. 
Indeed, about 360 a.d., King Crimthann ruled 
both Britain or Alba and Ireland. The most 
important colony came in 501 with the sons of 
Ere. The Scots latterly extended their con- 
quests south and north, so that when the Norse 
came in 794 they called the Minch the Scotland 
Fjord and the Pentland Firth the Pictland Fjord. 


Soon thereafter the Scots took supreme rule. 
The above are not Dr GiUies' views. Firstly, as 
to Argyll. He has created a new Irish word for 
the occasion; this is " oir-thir," East-land, from 
" air" or " oir," before, on, and tir, land. 
Now there is an old and a modern Irish word 
like this — " airther," the east, front part; it 
comes, as Dr Whitley Stokes points out, from 
the comparative of air — comparative in ter, 
Greek tero-s. He gives the old Celtic as 
(p)areitero-s, allied to Latin per, pro. Like 
English prepositional comparatives, it is used as 
a noun. It will be seen that Dr Gillies creates 
a new word in oir-thir, for tir has nothing to do 
with the Irish word. Besides the tir would pre- 
serve its long sound in the compound oir-thir. 
This argument topples one of the Doctor's card 
houses. Then as to his history. In Argyll, he 
says, Gaelic ' ' has been there from the begin- 
ning." Getting more poetical, after the Biblical 
manner — but somewhat after, he adds : — ' ' It is 
written in the rock." Now this same patriotic 
Gael allows that Eachairn, and especially Echdach 
(nominative Echaid), are the Gaelic descendants 
of Ptolemy's Epidii, the inhabitants of Kintyre, 
the p of which proves it Pictish or Brittonic ! 
Gaelic " each " is Welsh " ep," " eb." As a 
matter of fact the Gael did not visit the Epidii for 
at least forty years later. Argyll was then m 
the hands of the Picts, who spoke a Brittonic 


tongue. It is scarcely worth while noticing that 
he deduces Fergus Mac Erc's pedigree from 
Conn Ceudchathach, and not from his son-in- 
law, Conaire II., descended from Conaire Mor; 
but such are the unfailing facts according to tiie 
annals. He jvill antedate the coming of the 
Norse by two hundred years; why, one cannot 
see. 8ome harum-scarum youth lately an- 
nounced that a German professor held this 
belief; but when proof was asked in face of the 
overwhelming evidence on the other side, it was 
discovered that the ' ' Norsemen ' ' were the 
hired men that slew St Donnan and his 52 fol- 
lowers in Eigg or in Sutherland ; the annals call 
them "pirates," using the Latin term piraiti. 
The " Saint's Life " is responsible for a queer 
story of a queen taking vengeance on Donnan 
through hired pirates. The evidence that the 
Norse first came in 793 to the East Coast and 
burnt the great Monastery of Lindisfarne is 
firstly contemporary, and, secondly, Dr Gillies 
ought to know that what affects the Church for 
good or bad is sacredly recorded. " The Norse- 
men made a bee-line for the monasteries," once 
they discovered their wealth. The Norse ap- 
peared in Scotland Fjord in 794, and visited 
lona according to one account; in 795 they 
appeared on the Irish coasts ; in 802 they sacked 
lona; in 806 they slew the whole familia, 68 
souls. Dr Gilhes, of course, calls the pirates 


who killed Donnan "Norsemen" — no mistake 
about it. 

Other little foibles are there. The Scots are 
felt in modern Argyll as Easterlings, or Alban- 
aich even. He is unaware that Alba meant 
Great Britain till the 10th century (see for 
instance Cormac's Gloss, " Mug-eime"). The 
Druids, too, appear. Innis Drynich— he hesi- 
tates between " droighneach," thorn-wood, and 
" Druidhnich," Druids. The meaning of this 
last word, for it is genuine, " druineach," is 
artist or artificer, ornamentator. The name 
appears in Cladh nan Druineach (lona). Cnoc 
Druidean he corrects to Cnoc Druidhean, but 
Bishop Eeeves, who was there, makes it " Knock 
of Starlings " ! The Church part of the book 
is fairly done. St Finlagan he has missed; 
his chapel was in Island Finlagan, where tlie 
Lord of the Isles afterwards held high festival. 
This is all the more remarkable as Mr MacNeill's 
" Guide " tells all about " Sanctt Finlagane at 
p. 74. The name was used in a patronymic; 
Archibald M'Linlagan was at Stremnish in 1686; 
this is M'Gill Fhinnlagan. The name is a double 
diminutive of Finding, which the Scottish Gael 
corrupted into Findlaoch, whence Finlay. The 
funniest mistake is about isle Davaar. It is 
called the Island of " Sanct Barre," 1449-1508; 
the form Davaar is for older Do-Bharre, " thy 
St Barre," on the same principle as Mo-Barre, 


or Mo-luoc, etc. Dr Gillies devotes a paragraph 
to Davaar, and seems to accept the popular deri- 
vation "Double-pointed" isle. At anyrate, his 
history is at fault. Another saint with " do " 
has put him wrong; this is in Kildavie, which he 
renders as St David's. It is really dedicated to 
Do-Bhi, whom he knows as Mo-Bhi. There is 
another such in Skye, which the present writer 
also rescued from other saints. Mundu, as from 
Mo-fhindu, is good enough phonetics for Dr 
Whitley Stokes, and Dr Gillies need not boggle 
at it. Brannan is not from " bran," raven; the 
saint's name is really Brenaind, the Brianult of 
Martin. Maoldoraidh is a good name in itself, 
it is not Maoldeoradh. Maolrubha means 
" slave, or king of the promontory." Dr Beeves 
hesitates between " rubha," patience, or pro- 
montory. For examples of such names as Maol, 
with abstract, material or place-names, see 
Gaelic Society Trans. XX. There is a St Finan 
apart from the Findans ; he is in Ardnamurchan, 
Glengarry, and Abriachan. The root is " fin," 
shining, which appears in Glen Finain, or Glen- 
finnan; not fionn, as Dr Gillies has it. Where 
is Kilmodan explained? 

Looking at the district and island names we 
first find Dr Gillies shaking his head over the too 
easy derivations of Cowal and Lorn, from the 
names of the grandson and son of Ere. For the 
former he suggests, after much thought, the 


" feckless " idea of " comhdhail, " a meeting! 
Had he looked at Irish names like Fer-managh, 
Kinel-ea, or Iv-erk, he would see that these Vv^ere 
personal names originally. Men of Monach, Kin 
of Aodh, and O'Ercs, used now as land names. 
The ' ' Chronicles of the Picts and Scots ' ' makes 
it clear that these two names were originally 
Cinel Loairnd and Cinel Comgaill. The " Cinel" 
simply was dropped in course of time. The 
tract on the Scots of Dalriada shows other such 
" Cinel " names, and it is worth while examin- 
ing it to see if more district names might not be 
unravelled. Gigha isle, M'Vurich's Giodhaigh, 
he derives from Norse " gja," a chasm, bay, 
borrowed into Gaelic as " geodh," creek, and 
" ey," island, the whole being equal to Gja-ey, 
" rift island." The Norse called it Gudhey, 
God's isle. Dr Gillies knows better than the 
Norse themselves. Kilmaillie he renders into 
Gaelic as Cill A' Mhailuibh ; there is deep reason 
for the Gaehc article. Dr Gillies has evolved 
from his inner consciousness a set of " black 
friars" before ''friar" times, and calls them 
"mael," tonsured one, " dubh," black. The 
worst of it is that " mael," devotee, is never 
qualified by an adjective. When it is, the word 
is a confusion for " m41," prince, as Mael- 
mordha for Mal-mordha, "great prince," the 
name of some thirty kings and lords recorded m 
the ' ' Four Masters. ' ' Maeldub is ' ' Dark lord, ' ' 


and is the name of four saints. But Maeldub 
could never produce the phonetics of " Maili '' 
of KihuaUie. If Dr GiUies could assure us that 
Maldubh preserved its a, though changed to ao 
elsewhere, then we should allow his derivation 
from Maeldub, meaning " Black Prince." The 
Kvlmalduff cited bv Dr Gillies as the first form of 

*■' •/ 

Kilmaillie belongs really to Inveraray; the mis- 
take is quite inexcusable. Dalmaillie and Inver- 
maillie contain river names. It is a common 
word possibly from a Celtic " Madlios," root 
mad, wet, as practically the Doctor says, wilh- 
out, however, giving the root forms. 

Our author (Dr Gillies) thinks he has made a 
distinct hit in his derivation of ilrdnamurchan. 
The word appears in Adamnan (704) in the nom. 
pi. and dat. pi., thus: — Artda-muirchol and 
Artdaib-muirchol. The first part means 
'■'heights," not height (Dr G.); Reeves made 
murchol into ^' sea-hazels," and Bodlev's lib- 
rarian lately made it into " heights of the sea of 
Coll," which is not so bad. Our author at anv- 
rate accounts for the modern h ; he makes it to 
be Ard na mur(dh)ucan, "height of the sea 
nymphs"; the word murduchand means syren 
in early Irish, from muir, and duchand, singing 
(K. Meyer). The length of the u, we fear, spoils 
Dr GiUies' beautiful derivation; it won't leave 
the word in its most modern form. In Gaelic it 
gives three syllables. Besides, it does not a bit 


suit Adamnan's phonetics; he was most accurate, 
and the MS. was written only a few years after 
his death. He cannot be trifled with even over 
the / at the end. Rum seems to appear in con- 
nection with St Began, called "of Eumm" (676), 
Norse " rymr " won't do (Dr G.). Eigg is also 
in Adamnan, but the Doctor has no hesitation m 
taking it from Norse "egg," edge, which is 
absurd, as Euclid has it. Adamnan calls it 
Egea, possibly from " eag," cleft. It is so. 
Canna isle he deduces from " kunnu," know — a 
very poor " look-out " indeed, especially as to 
vowels. The word " cana " means porpoise in 
older Gaelic. Mull appears as Malaeus m 
Ptolemy, and so the Norse Miili, a point, is out 
of the question; Adamnan has Malea. The 
modern phoneftics are correct — liquid a short 
becoming ii (ball, buill). This really should be 
a lesson in rash etymologising, for even other- 
wise Muile, with its short vowel, could not come 
from Muli with long u. But Dr Gillies calls 
it the ' ' manifest derivation ' ' ! Truly with him 
" vowels count for nothing." Of course Colt 
(G. Colla) comes incontinently from the Norse 
— here from " Kolla," a hind. The minister of 
the island, who has made a most capable survey 
of the place-names, and who gave the results to 
the Inverness Gaelic Society, says he has found 
plenty hazel in the north of the isle, at the 
nearest landing point to the mainland. The 


name comes, as does Colonsay, from " Coll," 
hazel. Colonsay appears in Adamnan as Colosus 
— no n — and is pre-Norse. Our author lakes it 
from Norse " kollr," hilltop, another "feckless" 
derivation. lona and Islay possibly belong to 
the same root ; here we deal with ' ' funda- 
mentals." The root seems to be Celtic 
" (p)i," Aryan pi, pi, water, drink. We know 
it in the Esk rivers, Ptolemy's Iska. lona might 
be Aryan Pi-va, Pi-vi (locative); Islay, Pi-lia. 
Islay appears in Adamnan as Ilea, yet Dr Gillies 
thinks the termination is Norse " ey," island, 
and yet the Norse called it nothing but II, not 
Il-ey ! Such is the philology of imagination. 
Dr Gillies does not etymologise the roots of lona 
or Islay. We may add the Awe river, Adam- 
nan's Aba, which of course contains the root of 
" abhainn," nearly "naked"; the same root, 
with double stem, is in Avich. Dr Gillies was 
to explain it by a note to p. 58, but in this, as m 
other cases, there is no note. 

Some words our author has strangely missed 
the form and force of are these : — " Aoirinn," 
mass, the offerendum, whence Inchaffray, m 
1190 Inchaffren. It occurs in Rhu na h-aoirinn, 
Eilean na h-aoirinn, and Erin, Iring (Ardna- 
murchan). They are places where mass was 
held outside. The Doctor here suggests that 
very much over-worked Norse word " eyrr," a 
beach. The n he does not explain, for the word 


was never adopted into Gaelic, and the n cannot 
be the Norse article, which the Doctor does not 
seem to know about. " Longart," shielin£(, 
encampment, comes from longphort, originally 
meaning " harbourage"— "ship" — port. Tay- 
lor, the water poet, speaks of the hunting booths 
in the Grampians as Lonquhards (1618). Dr 
Gillies cannot explain it, since ' ' the supreme 
scholar of our time, not only of Gaelic, but of 
all languages, has failed with the word." The 
scholar meant we do not know; perhaps the 
Doctor speaks " sarcastic," but the word was 
explained in Inverness Gaelic Society Trans, 
fifteen years ago. " Lochay" is an unfortunate 
miss; Professor Mackinnon in 1887 explained 
the word as the translation of Adamnan's 
"Nigra dea," his Loch-dae in the index of 
chapters. The ending is the gen. of the old 
Gaelic word for " goddess." There are several 
rivers of this name all over Pictland. River 
worship was rife among the Celts as Gildas so 
tragically tells us. " Feoirlinn" is another word 
over which he hesitates, and at last he lands 
wrongly regarding " hnn" as pool, and " feoir" 
as fjara, ebb, of Norse. The word means 
' ' f orthing' ' or farthing — farthing land ; it is 
common all over the West Coast; its phonetics 
(N. fjordhungr) are represented by " birhnn," 
Norse "byrdhingr." " Doirlinn " is surely 
native. Elerig, lolairig, etc., of which some 


two Inindred or more occur, usually as Elrig or 
even Eldrig, is for " eileirig," an obsolete word 
meaning the place where the deer were driven 
into, a cul-de-sac, generally beside a hillock or 
hill, where the deer-slayers took their place. 
The word is in the " Book of Deer " (ind-elerc), 
now Elrig. This explanation has been public 
property for ten years. Glen Amamd should 
be compared to the famous Glen Almond; the 
river name Almond is good Celtic "Ambona," 
root amb, ab. " Leth-allt" is a burn with one 
high bank, for the word originally meant " cliff, 
height" (Lat. altus), and its Scotch use is due 
doubtless to Pictish. " Laimbrig," a landing 
place, has been explained as Norse " Hladh- 
hamarr," pier or landing rock, plus the word 
"vik," a bay (Gaehc Society Trans., XXI., 
317). " Corpach " is rightly explained first, 
but why fly to Norse " Korpr," raven? Are 
there really Norse words in Lochaber at all? 
We think not — at least not east of the Linnhe. 
"Corran" means a point, the fern, is " corrag," 
finger; it is common in the Isles; what has 
corran, a sickle, to do with it? The root of 
corran, sickle, is " kerp, korp," to cut (Stokes). 
Dr Gillies follows good company in etymologising 
Nant (better Neannt). It is an Englified form 
of the native rapid pronunciation 'n-ann-da for 
'n abhainn dubh, " Black river." With Coille 


or Drochaid prefixed the word is wonderfully 
" crashed " together. 

Dr Gillies has not done well in trusting to the 
ordnance maps. The gem of the book is p. 59 ; 
here on the Awe near its junction with the loch 
the map has "Conflicts," and below "1300- 
1308," referring, of course, to the feud between 
MacDougail and Bruce. The Doctor thinks the 
place is called " Conflicts," and gives a Gaelic 
" Coingheal " ! The map has Clenamachrie, 
Dr Gillies corrects to Gleann na machrach, but 
the true name is Cladh na Macraidh (Churchyard 
of the Young Men). The name Cormac on the 
same page is mismanaged; while another Cladh 
on page 58 is given as Cleugh and derived from 
Scots. Cluniter (51) is rendered Claon-leitir, 
whereas it is a sand-bank — Claon-oitir; Drum- 
synie, on page 52, is Drum-sineidh, not from 
sian. To take a place or two in Coll, for ex- 
ample — Airivirig is locally and by fact Airigli- 
mhaoraich, not from N. borg; Airinabost has no 
" har" in the name (shiehng-ton) ; Ascaoineach 
* is for Asknish (ash or ship ness) ; Clabhach stands 
foi' A' Chlabaich, and comes from clab — it has 
nothing to do with the words for ' ' kite ' ' ; Foill 
is not "treachery," but N. Fjall, hill; Gallan- 
ach is so named from a water plant of yellow 
colour growing there, and called by the natives 
gallan. The most extraordinary miss of all is 



in tlie case of Pharspig, Skerray — " I can make 
nothing of it " (Dr G.), but the local people know 
it to be from a very usual name for the sea-gull 
— " farspag." Loch Ghille-Caluim is really L. 
Cille-Chainnigh, from Kil-Kenneth, St Kenneth, 
where the church is. Glen Brander is from 
" Brandradh," ravens (abstract pi.). Tiretig- 
ean (p. 36) cannot be from Aodhagain or Egan. 
That he has trusted to these maps is also -proved 
by such remarks as at p. 72 — " I do not know 
the local history." 

As regards the Norse names, Dr Gillies gets 
more in love with Norse derivations as the book 

Proaig he will not have as Norse Broadbay, 
but it is pronounced like the Gaelic of Brodick 
save for the initial p, which is not likely to be 
Norse anyway. Gaelic often interchanges h and 
p in borrowed words; the dictionary amply 
pioves that. Crosprig and Librig show final 
" brekka," bank, or possibly "berg," hill, in 
the form ' ' brg . ' ' Melf ort is surely N . mela-f j ord , 
'■' bent firth," from the common word " melr," 
bent grass. Seil Isle is no doubt Norse; the 
Gaelic Saoil is phonetically the ideal form of 
Norse " seil." Soroba is the common Norse 
place-name " Saur-baer," mud or swamp or 
sour-ton. It appears as Sowerby in Yorkshire, 
Sorbie in Galloway, and two or three times in 
the Highlands. Gleann Fhreasdail and Loch 


Eestill are surely from ' Risdal," copse-dale, 
and not from " providence." Raoiceadal might 
be from " reykr," reek; Geodha an t-sil rather 
refers to N. sil, herring; Leodamus is Leod's 
moss (compare Skye, Strolamus, Skulamus, from 
Sturli and Skuh). Mi-mheall is possibly " mjo- 
fjall," narrow hill. Ernach is certainly not from 
" eyrr " ; the termination is Gaelic. Beinn 
Thuncairidh is likely Tunga-gerdhi, not sunna. 
There are many words, however, where Dr 
Gillies goes against the phonetic laws which 
govern the passing of words from Norse into 
Gaelic. Thus final or intervocalic kk, pp, tt, 
become simply k, t, p; k, t, p become g, d, b ; g, 
d, b get aspirated and disappear practically. 
Trodigal cannot be from trodhi; bodhi (sunken 
rock) is bodha; Lagal garve cannot come from 
lagr; this would be " lagh," as it is. Final rdh 
at the end of a second word becomes rd, rt, as in 
Suain-eart; " nd " also similarly becomes d, as 
in " miosad," narrow sound. Troternish is no 
exception, for the n of Throndar-nes is still heard 
and is preserved in old documents (M'Vurich, 
Trontarnis); we do not know if Trudernish is 
allied. Suaineart is given correctly derived 
from Sveinn-fjord, but at p. 11 Norse " jordii," 
our " earth," is introduced very unnecessarily. 
The word is not allied to Gaelic aird, pomt, which 
is alhed to Greek " ardis," of hke meaning. A 
bad blunder is taking Saddel from sand-dale ; the 


old form of the word would here save such non- 
seusp — Sagadull (15th cent.), M'Vurich's Sagli- 
adal. " Faodhail," a ford, is missed; it is from 
Norse " vadhill," a ford, a shallow between isles 
wliere horses pass (Miss Freer, " Outei- Isles"). 
Terminal -aidh in river names is Pictish; it 
is for -in, from -ios, -ia. Welsh represents this 
by -ydd; Pictish by -aidh or -idh, and Gaelic by 
-e. Terminal -an in river names has nothing to 
do with " abhainn " ; it stands foi' Gaulish -ona 
(Mairona, Divona — our Don), -ana. "Ard" is 
the Irish for " height," but Scotch GaeHc allows 
a locative according to locality. " Aoineadh " 
is a good Gaelic word, and cannot be from N. 
"enni." " Leac " is also a good word for 
'cheek," its locative being "leacainn," so 
commonly used for hill-sides. The Doctor says 
— " Leac, a check, a word with which I am not 
famihar " — just like his " leac " in making him- 
self arbiter of what is Gaelic, old or new (for its 
use in a school book see " Higher Gaelic Eead- 
ings," p. 78). Gleann-a-Comhann : the Doctor 
regards this a as the article; it is simply a glide 
vowel following n, and doing duty for the Irish 
eclipsis. Loch Sween and the old name Syiliii 
is from old Suibhne, a well-known name in Ire- 
land. Sweyn Mac Sweyn is also found in char- 
ters and documents (Coll register in Dr Johnson's 
tiiue; he spoke to Mrs M'Sweyn). There are 
three or four confusing surnames of this kind- — 


MacShui'ne (Argyll, M' Queen), M'Suain (Skve, 
M'Sveinii), M'Swan, M'Aoidhean (M'Quien, 
Skye) ; possibly MacCuinn from Conn ; but this 
last name is not found in Highland documents. 
It belongs to early history. Tormoid is from 
Thormund (final nd to d); Ivarr, lomhar, is for 
Ingvarr. Clan Ean Murguenich is surely the 
famous or infamous Maclans of Ardnamurchan. 
Dermot, son of Fergus Cerrbel, was the good 
King whom Ruadan cursed. Gometra stands for 
Godmundar-ey, " Godmund's Isle"; Hermitra 
for Hermund's. The name Oighrig or Eftric is 
not from oigh; the Abbess of Kildare had this 
name in 738, and it was spelled Aithbhric, later 
Africa. These are some of the errors which we 
find in Dr GiUies' Place-Names of Argyll. They 
are not all that we, or better still, one more 
acquainted with the county, could point out, but, 
as Mercutio says, " 'Tis enough." 



XoTB. — Tlie fstress acceut is indicated by a full fstop placed bet'ure ihe 
stressed syllable—, Aber.arder is stressed ou ihe third 
syllable ; .Ab( rscaig is streused on the first. 

abar, aber^ 43, 138, 331 
Aber.arder, 138, 208, 274 
Aber.ckalder, 43, 138 
Aber.deen, 43, 138 
Aber.feldy, 138 
Aber.geldie, 181 
Aber.nethy, 139, 320 
.Aberscajg, 8 
.Aberscross, 8, ix. 
Aber.tarf, 138 
abh, 331, xsvii. 
Aboua, 7 

Ab.riachan, 125, 138, 331 
-ach, 334 

Aclia.cliouleicli, 58 
Acliadli-da-tearnaidh, 58, 59 
Aclia.derry, 49 
Aclia.drom, 43 
Aolian.darroch,, 58 
Ach.bae, 60 
Ach..ducliie, 274 
Acli.eachfi, 60 
Ach.gourish, 319, xxTii. 
Acliiche, 60 
Ach. lean, 280 
Ach. loch, 24 
Aci.lum, 280 
Ach.luachrach, 49 
A oh. more, 58 
Achna.beachin, 265 
Achna.cloick, 60 
Achna.oochine, 181 
Aciina.coichen, 181 
Achna.hannet, 162 
Aohna.hiuich, 58 
Achtay.toralan, 59 
Adamrian, 33, 4?, ^7, 07 
Adru, 150 
Aebudae, 67 
Aedui, 2Q 

.Affric, Glen, 185 

-aidh, diminutive, 305 

.Aignish, 101 

ailean, 321 

.Ainort, liooh, 37 

Air, 107 

Airchartdan (.v. Ux-quhart), 

145, 152, 335 
Aird, 163 
airigh, 292 
Airina.bost, 353 
Airi.virig, 353 
.Airnemul, 170 
Airtkrago, 68 
.Aiadale, 91 
aite, 321 
Aith, 90 
Aithbhric, 35? 
Alasdair Car rich, 44 
Alba, 5 
.Alderney, 15 
Ald.larie, 269 
Ald.oairie, 140 
.Alisary, 291 
.Allan, 321 
allt, 140 

Allt a' Bhataich, 367 
Allt Buidhe, 244 
Allt Diiinne 'Ghoire Bkiu*, B»0 
A lit .Phearnagan, 245 
Allt .Gabhlach, 245 
Allt .Lairig, 243 
Allt .Lorgaidh, 245 
Allt .Lowrag, 244 
Allt .Mhadagain, 352 
Allt Ruaidh, 244 
Allt-saidh, 183 
Altana.bhraidh, 59 
Alta E-ipa, 7 
.Altavik, 36 
Altna.karrie, 308 



.Alter wall, 14 

.AltreboU, 14 

.Alvali, 156 

.Alvee, 156 

.Alvie, 156, 192, 235, 245 

.Alyth, 156 

an, 331, xxvii. 

Angii«, Eail of, 19 

Angus of the ielesi, 43 

Angus-Bone, 24 

.Anuat, 50, 162, 283 

Antiquities, 27 

Aoiueaclh, 356 

.Apavatn, 16 

.apdaint>, abthaiu^, 285 

.Apigill, 16 

apor, 43 

Aporioum Stagnuni. 42, 

.Appiu, 285 
.jVppIecrcss, 54 
.Ajrbol, 292 
.Ardacli, 59 
Ard.brylacli, 204 
Ard.elve, 60 
Ard.elester, 108 
Arder.sier, 127, 155 
.Ardincbe, 200 
Ardma.reo, 160 
Axd.meanagh, 22 
Ardna.imu'chau, 70, 145, 

Ard.narff, 58 
.Ardnish, 38 
Ard .Thuirinisb, 38 
Ard.verikie, 274 
Ard.Troilach, 269 
Ar.gyle, 64, 342 
.Arkaig, 180 
Arkle, 292 
.Arkkt, 180 
.Armadale, 14, 38, 179 
.Arnaboll, 11, 13 
.Arnamul, 171 
.Arnaval, 38, 171 
.Arnol, 98 
.Aruish, 38, 101 
.Arnisort, 101 
.Aros, 38, 104, 170 
.Arran, 70, 78 
Artbrananus, 32 
Artdamuirohol, 145 
-OB, 318 



Ascaoiiieacb, 353 

.Ascrib leles, 3G, xiv. 

i'^iiigrims-erg, 172 

.Ashore, 13, x. 

.Askary, 173, 290 

Asleif's Bay, Aeleifar-vik, 13 

.Asmigarry, 96 

Ash, 81 

.Assary, 290, 291 

.Assynt, 3, 6, 10, 11, 21, ix. 

Astle, 14 

Athole (Ath-Fhodhla), 5 

Auch. alter, 242 

Auchin. tore-beg, 48 

Auchter.chepar, 320 

Aiichtertyre, 59, 60 

.Aulafiary, 173, 291 

.Aulavaig, 37 

Auld.earn, 148, 331 

Aiiltna.sou, 62 

.Avernish, 57, xvi. 

Avie.lochan, 182, 272 

Avie.more, 182, 272, 331 

Avin.lochan, 182 

Awe, 350 

.Awloche, 24 

Ayre, 179 

Ayre Point, 38 


.Baccaskill, 82 

Bachd, 82 

Back, 82 

Bacca or Backa, 82 

.Backies, 13 

Bad Each, 251 

Bad Earbag, 250 

Baden, lougart, 175 

.Badenoch, 164, 186, 233, 259 

Badi.caiil, 61 

Bad na Deimheis, 256 

Baile-.chloichridh, 276 

Bail-in-tian, 175 

Baillie of DooMour, 205 

Bakka, 82 

Bakki, 13, 82 

Bala.chroan, 204, 269 

Balan.scrittan, 280 

Bal.chroan, 204 

Bal.dow, 264 

Bal.four, 125, 263, 334 

Bal.gowan, 271 



.Baiigiil, 16, xii. 
Balin.tore, 333 
Bali.viiilin, 262 
Bal.keith, 308 
Balla.cJuilia]i, 48 
Ballaii.truplial, 112 
Balleua.g.ulleich, 163 
Balli.feary, 123 
Balliii.luig-, 261 
Balliol, King Ed.,ard, 34 
Ball.misliag, 273 
.Balloch, Glen. 252 
Balma.carra, 54, 59 
Balnia.glaeter, ir^T 
Bal. moral, 182 
Balua.craig, 279 
Bal.naifl, 279 
Bal.nesjjick, 163, 205 
Bal.vamch, 163 
Bal.vatton, 314 
Imu (Welsh), 331 
.Banavie. 180 
Banba. 180 
Banff, 180 
Bamiockbu I'n , 25 
Bauquo, 43 

Barony of Gruids, 4, 8 
Barra, 74 
.Bajrapoll. 95 
Bari-c. 306 
Barro-vindo6, 306 
Barvaf^, 104, xxi. 
Bata.bog, 262 
Baude, 322 
.Beagram, 170 
Beallacli .Colluscard, 179 
Beallaoh na Sgaiide. 179 
.Beallid, 270 
.Beariaraig, 37 
Beauly, 10, 128, 1G3 
Beauly Firth, 6, 9 
Beinn, beann, 95. 331 
Beinu a" Sgatli, 38 
Beinue Bhuidhe ua Sroiue, 
Beiuu .Eibhinn, 254 
Beinu .Loyal, 16, 61 
Beinn Stack, 10 
Beinn Thuncairidli, 355 
Beinn .Valaman, 61, 255 
am Beitliigh, 279 
Bel.faet, 180 
Belle.ville, 265 
Ben Alder, 255 


Ben Arkle, 16 

Ben .Armin, 13 

Ben.becula, 75, 169, xix. 

.Benchar, 204, 205, 270 

Ben .Edair, 150 

Ben Griam, 13 

Ben .Eayaval, 16 

Ben Mac.dni, 59, 190 

Ben .Nevis, 47 

Benuie, 238 

Ben Eatk, 11 

Ben Storr, 38 

Ben Strome, 17 

Ben Tee, 176 

.Beolary, 173 

.Bernera, 74, 92, 168 

a' .Bhuidheanaich, 250 

.Biacbna-begg. 262 

.Bialaid, 158, 266, 270 

.Bigeary, 39 

.Biggary, 172 

.Bighonse, 18, xiii. 

Bigswall, 114 

.Birkisco, 38, 170 

.Blairgie. 182, 272 

blar, 140 

.Blarach, 205 

Blare.garwe, 60 

.Blargie, 182, 272 

Blar-na-leine, 54. 140 

Blar-na-cleireacli, 48 

Blar.onr, 49, 140, xxiii. 

.Blaven, 169 

Blingery, 291 

Blinkbonny, 322 

Boar, The, 255 

Bo.chrubin, 141 

Boece, Hector, 67 

beer, 88 

.Bogie, 238 

Bchnntin, 49, 141, xvi. 

.Boisdale, 178, xxvi. 

bol, 13, 14 

Bo.leskine, 141 

Bo.lin, 141 

Bolstadhr, 12, 79, 82, 84, 94 

.Bona, 126, 153, xxii. 

.Bonar, 7 

Book of Deer, 26 

.Boreraig, 37 

.Boreray, 74, 79 

borg, 86 

.Borgarey, 87 



BoTgie, 18 
.Borlum. 177 
Borna.ekitaig', 37 
.Borrobale, 14 
.Borrowston, 111 
.Borsam, 98 
Borve, Borr, 18, 85 
.Bosset, 17 
bost, 12, 37 
.Bosta, 83 
both, 141 

Bothwell, Morays of, 19 
.Boursa, 15 
.Boiiata, 82 
.Bowaett, 17 

.Bracadale, 34, 35, 155, xiv. 
Brae-ctat, 4 
Braeia.tra, 58 
Brae Locli.aber, 44 
Brae.i'iach, 190 
.Braga, 88 
.Braijor, 88, xx. 
Bran, E., 257, 258 
.BrRnderecraig, 37 
Brana.hiiie, 91 
Brannan, 346 
Brass, 218 

.Breabost, 39, 99, 171 
.Breaclet, 99 
.Breakachie, 205, 275 
.Breakness, 88 
.Breaquoy, 99 
.Breasclet, 99 
Brecacatli, 275 
.Brecbiu, 202 
Breck. Brekk, 38 
Breidharhlidh, 99 

Breidbhat, 117 
.Breinish, 102 
Brin, 213 

Broadford, ll9, xv. 
Broch.nain, 184 
.Brodick, 37 
.Brogaig, 37 
.Brogar. 88 
.Bronnacli, 27? 
.Brora, 17 
Brougli, 85 
Brough of Biri^a. SG 
Brow, 88 
Bn'i, 88 

Bruce, David, 34 
Bruce, Eob©rt. 34. 43 

Brue, 88 

Brugarth, 88 

.Brunary, 173, 291 

Bruthach uan Spardau. 2S8 

Bught, 36, 123 

Bul.roy, 280 

Bu.nachton, 141 

Burg or Borve, 39 

Burghead, 315 

Biirns, 81 

.Burra, 86 

BurroAV Head, 35 

.Burwick, 86 

.Biista, 82 

Buster or Bister, 12 

Bute, 78, XX. 


Cabraoh, 334 

Cadha an Fheidh L<>chain 

Ubhaidh, 252 
Caereni, 6 
.Oaggan, 280 
.Caiplich, 259 
Cairn. gorm, 140 
Cait, 5 

.Caitliness, 3, 9 
.Calavie, L., 57 
.Callwst, 83 
.Calder, 138, 241 
Caledonia, 142 
Calf of Mull, 77 
.Callart, 48, xvi. 
.Callater, 241 
.Callernish, 103, xxi. 
Call Ghaig, 258 
.Oilligarry, 172 
.Calva, 15 
.Calvie, Glen, 57 
.Camalaig, 37 
.Camas .Longart, 62 
Camerous, 25, 44 
Cameron-Campbell of .Monzie, 

.Campbeltown, 48 
C-amua-.Killean, 343 
.Canan, 214 
.Canna, 76, 349 
Can. tyre, 71 

Caochan a' Chaplioli. 259 
Cape Wrath, 6, 18. 68 
.Carbost, 171 



.Caiiehader, 108 

.Caxloway, 116 

cam, 140 

.Cariiaeh, 144, 305 

Cam ail Fhidlileir Lrorgaidh, 

Cam Dnbk Mhii au Deoir. 249 
.Cam«e«j 103 
.Garnish, 102 
Caimonacae; 6 
Carr.bridge, 321 
.Carrol, 17 
.CarroD, 144 
Castle Heather, 184 
Cat, 5, fl, 9 
.Cataool, 87 
Cataneeia. 4. 6, 19 
.Cataobh. 4. 5, viii. 
oathftii. .302 
.Catlodt'e, 275 
.Cattan. 208 
.CattanachB. 211 
Oatti, 5. 6 
Oausor. 320 
.Cawdor. 241 
C6. 6 

Cealiaii, 162 
ceantt, 3.-1 
CeaBn-i>i: -Oba, 67 
.Cearbbaig, 17 

Cfeleetine of Loohalah, 44, 6.3 
Cerones. 144 
.Cbaligaig, Port. 17 
Cbap«l-parfe, 267 
.Charlenton 117 
Cbattan, CJlan, 906, 224 
Chatti, 6, 7 
A' Chftoimich, 258 
Cheehohne, 24 
Chewil, 224 
A' Ohlabaich, 353 

ChlBanach, a'. 175 
Chrona. 16 
Oill-chro, 161 
rSlI 'DoTii'>nii!. 154 
CHlJ-DiMiin. 154 154 
Oirig, 6 

Oiffte .Mhearad, 260 
Clach an Tniiseil, 112, xxii. 
Clacb Mh-;. Cailin, 261 
Clachna.cauiin, 122, 332 
Oach »a h-Annaid, 162 

Claclina.liarry, 210, xxii. 
Cladh .Bhiallaid, 270 
Cladh Bhrighde, 270 
Cladh .Eadail, 270 
Clan.chamron, 224 
Clan.donochie, 224 
Clan Ean Murgiie ich. 367 
Clan-Guin, 21 
Clann Mhuirich. 226 
Clark, Alex., 24 
Clark, John, 259 
Clash.jiieil, 100 
.Clayside, 17 
Cleit, 10, 38 
Clena.machrie, 353 
Cleit, Ben, Loch, 38 
Clett. 99 
Cleiigh, 353 
clochan, 284 
Clon'.tarl, 70 

.CHiiainidh, 271 

Olune, 204, 205, 270 

Clunes, 24 

Clunie, Cluny, 175, 205, 271 
ClTiniter, 353 

Cluny Mao('hersoi)c>, i2f) 

.Clurie, 193 

Clyue, 11, ix. 

Cnoc Druideau, 345 

Cnoc .Prangach, 250 

Cnoc Praing, 249 

Cnoo nan Ajugeal, 61 

.Cnoideart, 79 

Coal, 99 

Cobhsamul, 80 

C^vilin.Uiie, 265 

(iciile.inore, 61 

.CoiLsay, 74, xix. 

.Coinnmheal, 95 

ooire, 140 

Coii-^-an-.Longairt, 175 

Coire Bhein, 254 

Coire Bhoite, 253 

Coire Bhran, 258 

Cfire Bog, 2.50 

Ooi re . Fhe a r u agan , 260 

Coire na Stoma, 62, 164 

Coire .Neachdradh, 251 

Coire .Phitridh, 254 

Coire .Seasgach, 57 

C-oire .Siiileagach, 255 

Coire .Yairack, 253 

.Colaboll, 14 



.OoUxwt, 171 

.Coldbackie, 13, x. 

.Colivick, 118 

Coll, 77, 99, 349 

.Colonsay, 68, 69, 78, 350, xviii. 

.Colsettr, 99 

Cohimba, 160 

.Cojiihraig, 245 

Ooinyn, 195, 196, 197 

.Conamheall, 15, xi. 

.Oouan, 334 

.Conchra, 60, xvi. 

.Consesaid, 17, xii. 

Conilicts, 353 

.Couista, 39 

.ooupairche, 284 

.Contin, 49, 280 

.Copeval, 96 

Oop-na-Hoe, 38 

.Ooraldie, 270 

Cor.arnstil-beg, 244 

-more, 244 

Cormac, 353 

Oornavii, 6 

Cornwall, 6 

.Corpach, 50, 180, 352 

.Corran, 46, 49, 173, 352 

.Corranaoh, 205, 270 

CJorry .Mhadagain, 140 

Corse, 100 

.Oorsopoll, 95 

Oor.iianan, 48 

Coulin.liun, 269 

Coull, 273 

Cowal, 346 

.Coylum Bridge, 181, xxvii. 

.Cracaig. 37 

Craig, 58 

Craig.ellachie, 182, 189. 249 

Craig Righ Harailt, 194, 232, 

Craig Ruadh, 255 
.Crakavick, 171 
.Crathie, 273 
Crathy, 205 
.Cravodale, 98 
Oreag an Abhaig ;>.' Bhail'- 

shioe, 252 
Creag an Loin, 262 
Crea.garry, 172, xxvi. 
Creag Bheag Chimi-a' ghiubh- 

saioh, 252 
Creag Chrooau, 256 

Creag Dkubli Bliiallaid, 253 

Oreag .Fhiaolach, 261 

Creag .Follais, 261 

Creag .GlnubhsaoLau, 260 

Oreag Lcathain(n), 260 

Qreia^ Liafch, a' Bliail'-shuari, 

Oi"eag .M.higeaohaid/1, 260 
Creag Mhor Bhail' *' olirotb- 

ain, 252 
Creed Rivei-, 8, 81, viii. 
Creicli, 11, x. 
(.'reouee, 144 
Cricli Cliat, 6 
.Crisigill, 97, xxi. 
Croft. carnocb, 265 
riroft.gowaji, 263 
.Orogary, 172 
.Ci'oigarry, 96 
(iroit-nio-Cliroetaii. 101 
.Ci'omaran, 279 
.Cromai-ty, 307 
CVose, 100 
Crosprig, 354 
Cross, 100 
.Crossapoll, 96, 100 
.Cros8l>ost, 83, 100 
.Cro^tsbreck, 38 
.Croesiix)!!, 95 
.Crossnish, 38 

Crossphiiil, L., 13 

Crowlista, 111 
Croy, 128, xxiii. 
•Oroyla, 259 
.Crubiu-more, 27*! 
.Oriiely, 118 
.Oniitlme, 6 
.Cruithuig, 6 
.Cruuuachau, L., 248 
.Cubbin, 29 
Cu.chulinn, 33 
Ciiid, 7 

Oiiidh.raug, 107 
Cuig.s of Strathdenr., 1«3 
Ouilc, an, 257 
Cml.chenna, 49 
Ouil-chuiuneig, 184 
Ouldees, 93 
Ciil.loden, 184, 335 
Cul.malY, 153 
.Culroes', 43, 335 
Culthnok, 60 
.Cumbraes, 78 



Oiimmin, Cuimiu, St., 128, lUi 
Cumuli nga, 43 
Ourr, 193 
.Ciirsetter, 107 
Cyder, hall, 13 

Dail au .Longaixt, 175, 275 

Dail ua Ceardaich, 276 

dal, 329 

-dal, 38 

Dal.aunacli, 276 

Da,.arossie, 127, 158, 335, xxv. 

Dal.chapple, 321 

Dal.cliully, 274 

.Dalcross, 156 

Dale.beg, 90 

Dale.more, 90 

Dales of Caitliness, 3, 9 

Dal. four, 174 

.Dalginrose, 166 

Dal Haxald, 12 

DaLkeith, 301 

Dal-.Langal, 17 

.Dallas, 319 

Dal.mally, 153 

Dal.migavie, 260 

Dal.moire, 109 

Dalna.vert, 203, 204, 279, ;^.:5 

Dal.navie, 308 

dalx, 13, 14, 15, 89, 109 

Dal.raddy, 205, 262 

Dal.uar, 334, xxvii. 

Dal.whlnnie, 275 

Da.vaar, 345 

.davach, 335 

Davidsons, 211, 226 

.Daviot, 127, 156 

Dean of Liemore, 33, 63 

Dean Munro, 72 

Dearc Beinne Bigfe, 254 

Deoantae, 5, 6 

Dee, 122, 147, 333 

.Deernese, 99 

Del. four, 125, 204, 205, 263 

.Deeher, 320 

.Deverott, 148 

Dicksone, 24 

.Diebek, 82, xx. 

Digg, 180 

diminutive -»idh, 306 

.Dingwall, 9. 82, 129 

Dingwall of Kilduu, 54 

.Diraclet, 99 

Dirie-Cliat, 4 

Divona, 122, 147 

.dobhar, 140 

Docli.four, 125, 174, 263 

Dooh.garroch, 174 

Don, 122, 147 

Donald Ba.lloch, 44 

Donald Dubh, 44 

Donnan, 344 
j .Dor&s, 126, 157. xxv. 
! .Dorlin, 178, 351 

.Doruocli, 9, 10, 23, 306, ix. 

.Dorrery, 173, 290 

Dorsum Britanuiee, 146 

.Dourag, 140 

.Drocliaid Charra, 321 

Dronside, 12 

Drostau, 161 

Druie, 238 

druineacli, 300 

Drum. ar bin, 49 

Drum.buie, 61 

Drum.cliardine, 152 

Drum.gask, 271 

.Drummond, 123, 175 

Drum.ochter, 255, 257 

Drum.synie, 353 

.Duaircl, 61 

.Dubec, 82 

Dubh-Ghiubliaie, 192 

Duf, 24 

.Duffue, 19, 319 

.Duible, 14, x. 

Duirinish, 34, 35. 56, 78, 99, 

dul 139 

Dull, 139 

.Dulnan, 239, 335 

Dun.achton, 194, 26^ 

Dun Borgh, 86 

.Duncansburgh, 48, W-i 

.DuncauBbay Head, 7 

Dun.oraig, 58 

Dun-dearduil, 49, xvi. 

Dungeness, 100 

Dun.keld, 142. 176 
Dun.lichity, 158 
Dun .oily Caetle, 32 
Dun.robin, 29 
Diin-Scathaich, 33, xiv. 
Dun.stafnage, 180 



Dun.iuim, 36, 180 
Dun.vorrerick, 74 
.Diun^^B, 3, 6. 8, 10. 11, 21,56, 
78, 156 

Durria, 126. 157 

[hwarv. 173. 291 

Duthii, 167, 320, xxv. 
Dyksouit, Dickson. 26 
Dyx, 99 
.Dyrjk-kVttT. 99 


.Eabt**, 171 
Earn. 148, 335 
.BaarBhadei, 107 
Easga iui Lochain, 250 
EaMer Hoes, 10 
Elxmdae, Ebiidae. 145 
Eb\ida. 14A 
Etl»T. 91 

Eddra.ehille6, 10, 11, 15, 21 
Ediubnrcrh MS.. 54 
Edin *»tr 267 
Egea. 68. 145. 150 
.EidaTt. 246 
Eidird. 267 
Eidb. 90 

Eigg. 7e. 145. 150. :349. xxIy. 
Ei1(^, L.. 47 
Eilean an Rigli, 247 
Eileai) Bwigram, 170 
Eilenn H«ist, 179 
Eilefttj Han Oon, 247 
Eitean Tioram, 62 
.Eing-an#«j 99 
eirbhf, 308 

.ELsbort, Locb, 37. xv. 
Eist 179 

Ekkjalbakki, 9, 13 
Elans hoga, 15 
Elane.neyfe, 15 
Elane.qnothra, 15 
Elane.Ton*, 15, 15 
Elan.gelye, 15 
Elaji.nmnne, 46 
Elan.T?illijrh«>. 15 
.Eldrabk. 14, xi. 
Elena. 68 
.Eljrin, 322 

Elg^Kf•hire, Place Names of, 

.Ellishader, lUt! 

.Elliater, 108 

.Elrick, 176, 253, 351. ixvi. 

.Embo, 11, 13, X. 

-Enaclete, 99 

.Enard, 93 

Endi, 91 

.Ensay, 78, 92, 168 

.Eoradale, 90 

.Eoropie, 88 

Epidii, 69, 343 

Epidium, 67, 69 

.Eport, 93 

.Erbusaig, 56 

.Ercliless, 185 

.Erebie, 107 

.Ereboll, 11, 95 

.Erichdie W Peter, 248 

.Ericht, Loch, 248 

Erin, 350 

.Eriskay, 74, 169 

.Erisort, 93 

.Erista, 112 

Ernach, 355 

.Erraid, 77 

.Erribol, 13 

.Errocht, 50 

.Ersaiy, 291 

Esk, 332 

.Efekadale, 10, 167, 306 

.Essich, 183 

Esain.tiillich, 181 

.Eswick, 90 

.Etteridge. .Ettridge. 181, 20.5 

Ey, 15, 91 

Eye, 91 

.Eynort, Locb. 37 

Eyre, 38, 179 


Fadauiine, 60 

faithir, 308 

.FaLside, 17, xii. 

Faodliail. 356 

.Farigag. 126, 173 

.Farleitir, Farletter. 163. 279 

.Farmheall, 16 

Parniia, 157 

Parqulxai-.sons, 206, 211 

Farr, 11, 279 

.Farraline, 163 

.Fan'ar, 143 



Fassi.feru, 50, 180 

•Fearna, 240 

Fea.vorar, 262 

.Feithlinu, 3S0 

fell, 79 

Fell, 95 

feodhail, 308 

Feoirliun, 351 

.Fercliard, 215 et seqq. 

.Ferchard Leche, 15 

.Ferchar Fota, 208 

Fergus mac Ere, 344 

.Fernaig, 58 

Feruaig-beg, 60 

.Fernsdale, 244 

.Fersit, 49, 180 

.Feshie, 190, 208, 237, 238, 257, 

.Fia.ray, 75 
Fib, 5 
Fidach, 5 
Fiddes, 24 
Fife, 5, 20 
Finan, 161 
.Findliorn, 148 
Findlaoch, Fiulay, 345 
.Fingask, 271 
.Finnbar, 154 
.Finnlarig, 176, 321 
.Fintag, 47 
Fionnlarig, 321 
.Fiskavaig, 37 
Fivepenny Ness, 101 
Fjall, 15, 95 
Fjordhr, 15, 79, 93 
.Fladda, 71, 76, 168 
Flannen Islands, 74 
Fleet, 8 
• Flicliity, 156 
.Flodeway, 117 
.Flodigarry, 172 
Foill, 353 
Forbeses, 25 
.Fordun, 54 
Forfarshire, 21 
Forr, 279 
.Forre», 21d 
FoTS, 38 
.Forsan, 38 
Forse, 38 
Forsin.ard, 18 
Fort-Augustiis, 128, 183 
Fort-George, 183 

Poi-tr«un, 5 

Fort-Williaiu, 48, 183 

Fotla, 5 

Foula, Fula, 75 

-four, 334 

Fowlis, Lord Gray of. 21 

.Foyers, 128, 183 

.Fresgill, 16, xii. 

Preskin, 19 

.Fre.ucliy, 54 

.Puday, 75 

Fuidhaidh, 75, xix. 


Gaick, 256, 257 

.Gailval, 14 

Gairidli-Ghiumaig. 172 

Gallanach, 353 

.Gallaobk, 4, viii. 

.Gallovie, 182, 272, 321 

.Galloway, 25 

Gal-neap, 97 

.Galtrigill, 97 

Gar, 24 

Garbh-Ghriochan, 151 

Garbh-Gtaig, 258 

.Garbost, .Garrabost, 83, 332, 


Gardhr, 96 
.Gargask, 271 
. Garni orau, 135, TSl, 165 
Garra.fau, 172 
.Garragarth, 83 
.Garraquoy, 83 
-garry, 39, xv., xx 
Garry, 166, 238, 335 
Garry. more, 172 
.Garten, 175 
Gartnait, 32 
.Garva-beg, 27? 
.Garva-more, 273 
Gask, 204, 205, 271, 335 
Gask-beg, 271 
Gask-lyne, 205 
Gaslv-more, 271 
Gaudie, 23S 
.Gaulisli, 6 
.Gealcliarn, 140 
Geara-du, 172 
.Gearnsary, 291 
gearrairth, 287 
Geary, 172 




Qeirabolstadbr, 83 

.Geldie, 181, 238. 331, xxvii. 

.Gellovie, 206 

Geodha, 10 

Geodha an t-Sil, 355 

• Gesbader, 108 
.Gheallaidh, Allt, 181 
Ghoibhuidh, Allt, 241 
Gibbon, 29 

.Gigha, 78, 347 

Gil, 16, 38, 97 

Gilbert, Bishop of Caitliness, 

Gill.anders, 54, 211 
Giol, 97 
Girsa, 98 
.Girsigarth, 98 

• Gisofill, 16 
Gisl. 97 
Gja, 16 

Glac an t-Sueacbdaidh, 251 

.Glastor, 50 

Gleann Fhreasdail, 354 

Gleu Amaind, 352 

Glen Brander, 354 

Glen .Calvie, 57 

Gleu.carnie, 193 

Olen.coe, 334, 356 

Gleu-Couvinth, 185 

Glen.dessary, 48 

Glen.elg, 157, 322 

Glen.fesbie, 190, 257 

Glen. finnan, 178, 346, xxvi. 

Glen.fintaig, 47 

Glen.garry, 54, 96, 166 

Gleu-geoullie, 181 

Glen.gloy, 48 

Gleu.kingie, 48 

Glen.luy, 44, 48. xvi. 

Glen .Moriston. 54. 163. 1«6 

Gleu-ose, 37, 170 

Glen-pean, 48 

Glen-quoich, 180, xxvi. 

Glen.roy, 48 

Glen..shiel. 294 

Glen-sbirra. 273 

Glen-sulag, 48 


Glen.truim, 205 

Glen .Udalan, 61 

Glen .Urqnhart, 54, 152. 186 

.Gnipa, 97 

Goatfell, 95 

.Golspie, 10, 24, 46 
Golspie Stone, 27, 28 
.Golvall, 17, xiii. 
G^metra, 74, 357 
Gordon, 20, 21, 48 
Gordon, Sir Robert, 4, 13. 20, 

21, 22 
Gordons, 202 
Gordonsburgh, 48, 183 
Gordons' Hall, 277 
Gordouston, 20 
Gortincrief, 263 
Gow, Dnncan, 257 
.Gradsary, 291 
Graff nose, 98 
■ Grafirdale, 98 
Grampians, 189 
.Granish, 182, 314 
Grant, Mrs. of Laggan, 272 
Grants, 54, 204 
Gras, 98 
.Grasabbaig, 98 
.Grasgo, 170 
.Grassfield, 08 
.Grassholm, 98 
Graupins, 142 
.Graveland, 98 
.Graven, 98 
.Gravir, 98 
Grawine, 98 
Gray, 21 
.Greenwall, 114 
Greepe, 97 
.Gremiston, 110 
.Gresbornish, 38, 104, 180 
.Gresmark, 98 
Gress, 98 
Greta, 8 
Griam, Ben, 13 
.Griamacbarry. 13 
.Grimagarry, 172 
.Grimisay, 74 
Grimr, 13 
Grims-argb, 290 
.Grimsbader, 106 
.Grinez, 100 
.Griomabbal, 108 
.Grisbernisb, 38, 104. 180 
.Gi'odsary, 18 
Grof, 98 

.Grudie, River, 8 
Gniids, Barony of, 4, 8 
.Gruinard, 93, 94 



G rum-beg, 18 
Grumb-more, 18 
.Guinag, 240 
• Giilbin, Locla, 47 
.Gunna, 77 
Gunns, 21, 22 
.Gurshader, 108 

Ha or Hay, 225 
.Habost, 84, xx. 
.Hacklete, 99 
Haco, 12, 13, 33 
.Hacoin, Lochan, 12 
.Halikeld, 177 
.Halladale, 14, xi. 
.Halsary, 173, 290 
.Hamara, 38 
.Hamarr, 38 
.Handa, 15 
Harestanes, 317 
.Harport, 37 
.Harrapool, 95 
.Harris, 73, 159, xviii. 
.Harrow, 290 
.Hasker, 75, 109, xix. 
Hatti, 5, 7 
Hatton, 316 
Haiigh, 122 
Haversay, 76 
Heast, 37 

Hebrides, 9, 32, 67, 145 
Hebudes, Haebudes, 145 
Hecla, 169 
Heimr, 98 

.Heisker, 75, 178, xix, 
.Heistamul, 170 
.Helaval, 38, 169 
.Hellipoll, 95 
.Hellisay, 75 
-Helmsdale, 12, 14 
Helmsdale River, 7 
Herishader, 171 
-Hermetray, 74, 92. 357 
.Herries, 73 
Hesse, 5 
Hestam, 170 
Hig-h Bank, 7, J2 
Hillam, 18, xiii. 
Hinba, 68 
Hirt, 73, xviii. 
Hoe. 38 

.Hogi^etter, 84 

Holl, 98 

Holland, Hallandi, etc., 84 

Holme, 122 

Holm Isle, 36 

Homildou Hill, 203 

.Hoiiary, 291 

Hojie, Ben and Loch, 18, 57 

.Horgibost, 84 

.Horisary, 173, 291 

.Horneval, 38, 169 

Horse Isle, 77 

.Hon by, 84 

.Hougheary, 172 

.Houllaud, 84 

Hourn, Loch, 62, 164 

.Housay, 104 

.Hcusby, 104 

.Howbister, 84 

How. more, 170 

.Hoxa, 91 

Hoy, 74 

.Himish, 38 

.Huutly, 202 

Huntly, Earl of, 20, 44, 45, 49 

.Husabost, 39, 104, 171 

.Hushiuish, 104 

Huxter, 84 

hybrids, 332 

.Hysbackie, 13, x. 

Tarlraig, 253 
I Cataib, 6 
.Idrigill, 38, 179 
Ila fliimen, 7 
.Ilidh, 7 
.Inaclete, 99 
.Inchard, 15 
luch.fuir, 174 
Inch.nairn, 58 
.lugasteu, 99 
.lugyebiister, 99 
Inner. geldie, 181 
Inner. gelly, 181 
Inner. roy, 49 
Innes, Cosmo, 9, 22 
Innes or Inch, 49 
Innes, Sir Hugh, 55 
Innis Drynich, 345 
Insh, 204, 236 
Insh.navie, 285 



lnver.aro&, 38, 178 

Inver.eshie, 205 

Inver.lochy Castle. 43 

laver.markie, 203 

Inverna.havou, 200, 205, 27G 

Inver.uess, 9, 122, xxii. 

luverness. Place Names of, 133 

Inver.tromie, 205 

Inver.iiglas, 175, 278 

lona, 67, 68, 69. 350 

Ireland, 9 

Iring, 350 

Irland, 107 

Irt, 177 

.Isay, 36, 76, 168 

Iska, 332 

Isla R. 7 

Islay, 68, 71 

Isle-Colm, 15 

Iturnau, Itarnan, 309 

.iubhrach, 300 

James IV., 34 
Jolin of Isla, 43, 44 
Jolin of the Isles, 218 
Jiira, 15, 78 


.Kallin, 162 
Kata, or Katar, 5 
Keanin.tachair, 262 
.Keoldale, 14, x. 
Keose, 99 
Keotha, 99 
.Keppoch, 44, 49 
Keppoclimuir, 265 
.Kerrera, 77, xix. 
.Kerriwiek, 118 
.Kcrrow, 268 
.Kerwick, 17 
.Eessoc, 161 
.Kessock, 161 
Kil.aulay, 161 
Kil. bride, 160 
Kil. coalman, 161 
Kil.chattan, 210 
Kil.choan, 160 
KiLchiximen, 181 
Kil. Christ, 161 
Kilda.vie, 346 
Kil.donan, 11, ix. 

.Kildwick, 177 
.Kilham, 177 
Kill.eainan, 21, 309 
.Killegray, 75, 168 
Kill.ianan, 161, 178 
Killie.huntly, 278 
Kill.iu, 154 
Killochir, 58 

Kil.malie, 24, 46, 15-!, 317. 
Kilma.luok, 155 
Kilma.ree, 160 
Kil. martin, 161 
Kilmol.ruy, 160 
Kil.michael, 161 

Kilmo.nivaig, 46, 


Kil.morack, 154 

Kil. more, 154 

Kil.miiir, 35, 155, 160 

Kil.pheder, 161 

Kl.ravock, 158 

Kil.tarlity, 155 

Kil.vaxter, 154 

Kil.vean, 124 

Kin.cardine, 127, 192, 335 

Kin.cliyle, 261 

^<va crai^, 204, 264 

King.iLssie, 157, 190. 192, 236, 

Kinloch.more, 48 
Kin.mylies, 123 
Kinna.moine, 62 

Kn.rara, 203, 204, 205, 261 

Kin.tail, 53 

Kin.tyre, 48 

.Kirkaig, 17 

.Kirkapool, 94 

Kirk.hill, 157 

.KirkiboU, 14, 94 

.Kirkibost, 162 

.Kirkton, 61 

.Kirvig. 118 

.Kirkwall, 114 

Kjos, 99, 38, 99 

.Knappacli, 277 

Knock. ando, 321, 333 

Knock. sfranisb, 314 

Knock. navie, 285 

Knocknin.galliac'h, 262 

.Knoydart, 93, 166 

.Kolgrave, 98 

Kollr, 99 

.Kraiknish, 38 



Kroas, 100 
Kyleakiu, 13, 3* 
Kyle of Lochalsh, 61 
Kyle-strome, 17 
.Kyllachy, 182 

Lachan.dhu, 266 

.Lacsabhat, 117 

las, 318 

•Laggan, 158, 204, 237, 247, 268 

Laimbrig, 352 

Lairg, 3, 8, 11, ix. 

Lairig-dhru, 176 

•Laival, 169 

.Lamalum, 170 

.Lamigo, 16 

.Lampay, 37 

.Langaig, Loch, 37 

•Langary, 291 

.Langavat, 117 

.Langdale, 14 

.Langwell, 17 

.Larbert, 140 

Largs, 34 

•Laufafell, IG 

.Lawman, 82 

.Laxa, 81, 90 

.Laxdale, 89 

.Laxfirths, 90 

.Laxford, 15 

.Layaval, 169 

Leac, 356 

.Leachtin, 123, 175 

.Leanachau, 49 

•Leault, 264 

.Lecropt, 154 

Lee, 178 

.Leiphen, 174 

.Leirable, 14, x. 

•Lentran, 152 

Leodamus, 355 

.Lerwick, 14 

.Leslie, 24 

Lethallt, 352 

Letter.four, 125, 263 

Leum na Feinne, 256 

Leum na Laracli, 257 

.Leurbost, 83, xx. 

Leven, Loch, 47, xv. 

Lewis, 8, 10, 48 

Lews, 70, 73 

.Liaiiimul, 170 

Li brig, 354 

Lillingston, 55 

.Linshader, 108, xxi. 

Ling, E., 62 

.Lingay, 75, 169 

.Linmul, 80 

.Linshader, 107 

.Linside, 17 

.Lionel, 98 

Lis.more, 77 

.Liuravaig, 37 

Loch.aber, 42, 43, 48, 146, 151 

Loch.alsh, 53 

Loch an Diiin, 246 

Loch an Eilein, 245 

Loch an Laoigh, 57 

Lochan na h-Earba, 248 

Loch an t-Seilich, 246 

Loch Arkaig, 44, 47 

.Lochay, 147, 351 

Loch Broom, 54 

Loch .Calavie, 57, xvi. 

Loch .Carron, 6 

Loch Cille Chainnigh, 344 

Loch .Crunachan, 248 

Lochdae, 42, 147, 194 

Loch Eilde, 47 

Loch.gelly, 181 

Loch .Gulbin, 47 

Loch Hoiiru, 63, 164, xvii. 

Loch.iel, 44, 45, 46, 213 

Loch Insh, 194, 245 

.Lochlaun, 190 

Loch Leven, 47 

Loch Linnhe, 46 

Loch Long, 62 

Loch.maddy, 178, xxvi. 

i.och Ma.ree, 160 

Loch .Monar, 62 

Loch na h-Onaich, 62 

Loch.nellau, 321 

Loch Ness, 47 

Loch Oich, 180, xxvi. 

Loch Eestill, 354 

Loch .Sandwood, 117 

Loch Stornua, 116 

Loch Sween, 356 

Loch Tr6ig, 47 

.Lochty, 320 

.Lochy, 42, 46, 147, 3'JO 

Logachnacheny, 200 

Logi, Lougoi, 7 



ujgie-tevi, 140 

IjoiliUd Urbicus, 315 

Lon-biiidhe, 61 

Loudon, 56 

.Lionga, 68 

.Lougart, 275, 351 

Liong Island, 70 

Long, Loch, 62 

Loogdae, L., 194 

Lord of the Isles, 43, 44, 45 

Lorn, 346 

Lossie, 238, 315 

Loth, 7, 11 

Lovat, 185 

.Loyal, Ben, 169, xii., xxv. 

Loyninriach, 262 

L;ib Mhairi, 232 

Luga, 7 

Liigi or Lou^i, 6, 7 

Liiing, 68, 77 

.Lunasting, 28 

Liinda.vra, 49 

.Liindie, 56 

.Lunga, 68, 77 

.Lusta, 39, 267 

Lyn.chlaggr,u, 278, 279 

Lyne, 319 

Lyn.wilg, 205, 261 


Macbeans, 207, 211 
Macbeth, 9 
McCraithe, 24 
McCulloch, 24 
Macduff, 215 
McGhie, 25 
McGille.mor, 24 
McGill.espy, 2^? 
MacGillivrays, 211 
Macgill.ony, 225 
Mclvers, 25 
Mackay, Mr John, 18 
Mackays, 3, 21, 22. 25 
MacKays of Kintyre, 26 
MncKciizios, 2"!. 54, 55 
Mnckic, 25 
Mackintash, It. 
Mackiutoshcs, 206, 211, 214 
Maclvintoshes, 25 
MacKostie, 161 
M:;ckphail, 24 

McKraith, 24 
McKwatt, 24 
Maclean of Coll and Luchbuy, 

44, 45 
Maclean of Duart, 45 
Macleans of Dochgarroch, 211 
MacLeod, 8, 34 
MacLeods, 25 
McLinlagan, 345 
MacMahon, 23 
MacMurquhe, 25 
Macphails, 211 
McPhails, 22 
Macpherson, 25 
Macpherson, 262 
Macpherson, Lachlau, 242 
Macphersons, 211, 225, et sr(jq. 
Macqiieens, 211 
McKaithe, 24 

McSuaiu, McSwan, &c., 357 
MacVnrich, 11, 33 
.Maehall, 38 
Magh.rath, 11 
Magnus Barefoot, 32, 70, 71 
.MaTikel, 18 
.Maillie, 153 
Malaeos, 67 
Malag, 37 

Malmoran of Glencarnie, 203 
Malbrigd Bucktooth, 13 
Ma Luac, 267 
.Manachainn, 285 
Man and the Isles, 33 
.Mangarsta, 110 
.Mangaster, 110 
.Manish, 38, 103 
Manxmen, 71 
Maol.doraidh, 346 
Mao-rubha, 160, 346 
.Maraig, 117, 171 
.Marishader, 39, 171 
.Markie, 237, 243 
Markinch, 122 
.Marrel, 17, xiii. 
Martin, 72 
.Marwick, 117 

Mnryburgh, 48, 183 

Mashie, 237, 242 
Maw. more, 45 
Mathan, 23 

Mnthoson. 23, 54, 62, 209 
Ma.thesoii, Sir Keiinelh, 55 
Maxwell, Sir Herbert, 29 



.Meadak, 38 

Meadowside, 2G5 

.Mealista, 110 

Meall Aillig, 258 

Meall an .Dubh-chadha, 260 

Meall .Dubhag, 260 

Meall Horn, 16 

Meall na b-Uinneig, 253 

Meall .Rinidb, 16 xi. 

.Meanisb, 38 

.Meavig, Meavag, 118 

Mela, 67 

•Melbost, 83, 111 

.Melby, 111 

Melfort, 354 

.Melness, 16, sii. 

.Melsettr, 111 

.Melvich, 17, xii. 

Mbeall, 95 

Micbin, 29 

Middleton, 308 

.Migdale, 14, x. 

Mi-mbeall, 355 

Mincb, 177 

.Mingarry, 166, 172 

.Mingay, 36, 166 

.Minginisb, 34, 35, 36, 166 

.Mingulay, 75, 166 

.Minigaig, 258 

.Modsary, 18, 291 

.Moidart, 14, 93, 165 

.Moireacb, a' .Mbormlioich, k',;.;) 

Mol.inginisb, 104 

-Monadh, 140 

Monadli Caiplicb, 259 

Mouadh-liatb, 140, 189, 249 

.Mouar, L., 62, xvii. 

Monitcarno, 193 

Mouivaird, 217 

.Monkstadt, 110, 154, 163 

-Morar, 166 

Morav, Murray, 18. 19. 20, 21, 

24, "43, 322, 333 
Morgan, 26, 214 
Morile, 182 
Moriui, 322 

Mormaer of .Moyra. 11 
Morrisons, 25 
.Morsaig, 37 

.Morvern, 151, 165, xxiv. 
.Mosgaraidb. 172 
Mos-send, 91 
Mowate, 24 

Mowatt, 25 
•loy, 50, 127, 158 
Moyra, 11 
.Mucarnaicb, 30b 
.Muccoul, 274 
Muck, 76, 159 
.Mucracb, 180 
.Mudale, 14, xi. 
Mudy, 24 
.Mugeary, 172 
Mull, 100, 349, xxi. 
Mull. 68, 71, 77, 100 
Mul.roy, 44, 229 
.Mundale, 317 
.Mundwell, 317 
.Musal, 17, xiii. 


Nairn, 128, 164, 333, xxiii 

Nant, 352 

.Naver (Nabarus fl-am^ny. 4 

.Navidale, 15, si. 

.Navity, 285, 308 

Neave, 15 

Nebis, 149 

Neep, 97 

.neimbeadb, 284 

nes, 16, 100 

Nesa, 145, 146 

.iSlesbustar, 84 

Ness, 4, 122, 185, 332 

Nessa, 122, 146 

NessiutuUich, 181 

Nestos, 122, 146 

Nevis, 47, 149 

New.more, 308 

Newton. more, 269, 332 

Nicbolson, 27 

Nicolsons, 54 

Nigra Dea, 42, 146 

Nipa, 97 

.Nisabost, 84 

Nitb, 139 

.Nonach, 62 

Nona.kil, 285. 308 

Norse, 6, 7, 8, 10, 11, 32, 5G 

Norse-Gaelic pbouetics, :155 

Norsemen, 9, 69, 79 

Norse names, extent of, 70, 

Norse terms, 168 
Norway, 33 



.Nostie, 59 
.Nunton, 163 


Oakculiead, 317 

Ob, 18, 57 

Obau, 18, 57 

Ob-an-Duinc, 57 

Obbc, 57 

.Obisary, 173, 291 

.Ochil Hills, 9 

.Ochiltree, 9 

Og-ham letters, 27 

Og-ston, 317 

Oich, L., 180, XXVI. 

Oidecha, 68 

Oighreag, 357 

Oikel (v. Oykel) 

Oirir Gliaidlieal, 43 

Oiseval, 79 

.Oisgill, 37, 97 

Olave, 18, xiii. 

.Oldaney, 15 

Oldney, 15 

.Oldshores, 13 

Ommou, 68 

.Onicb, 49, 62 

.Oransav, Orousa, etc., 36, 78, 

89, 169 
.Orbost, 171 
.Orchy, 331 
Ord, 3, 7 
Ord Bain, 182 
.Oreval, 171 
Orkas, 7 
Orkney, Orcades. 7, 9, 32. 68, 

.Ormiston, 317 
.Orton, 317 
.Osdal, 37 
.Osdale, 15, xv. 
Ose, 37, 107 
.Oshmigarry, Osruig-arry. 96, 

.Oskaig-, 37, xv. 
.Osmnnd^vall, 114 
.O.siiaTrall, 9P 
.Osi3isdale, J 
Oss; 104 
.O.stai.g-, 37 
.Overton, 31J 

.:;,ic'. 1!7U 
.Oykfil, 3, H, 

.Pabay, .Pabbay, 36, 74, 93, 

Paible, 163 

Pairc-au-t-seipeil, 162 
.Palascaig, 61, xvii. 
.Papa, 93 
.Papadill, 105 
.Papal Burra, 105 
.Papa Little, 105 
.Papal Tell, 105 
.Papa StoTU', 105 
.Papa Stronsay, 105 
.Papay, 105 
.Papdale, 105 
.Papi, 104 
.Papley, 105 
.Papyli, 105 
.Partick, 140, xxiii. 
Paton, 29 
Patt, 62 
.Pattack, 243 
Paul Mac Tire, 22 
Payble, 105 
Pean, 43 

.Peffer, 129, xxiii. 
Peina.ha, 174 
Peiua.vaila, 174 
Peiu.chorran, 173 
Peiu.ess, 174 
Pein.gown, 174 
Pein.more, 174 
Peni.filler, 174 
Penin.erin, 174 
Pen.soraig, 37, 174 
.Pentland Firtb, 5 
Perth, 140 
Pet, 7 

Pethgnidie, 8 
Pettakarsie, 8 
Pettland, 5 
Petty, 126, 138. 215 
Petty. vaich. 137 
Phar.spig- Skerray, 354 
Phoine.«i, 181. xxvi. 
Plir>ne.s, 205 
P Irtish . 137 
Picts, 5, 7, 27. 43, 56, 207-8, 3J9 



Pit.airlie, 322 

Pit.almit, 137 

Pit.chalman, 137 

Pit.chirn, 137, 205, 208 

Pit.cliuTn, 263 

Pit.four, 8, 126, 174, 263 

Pit.fur, 174 

Pit.grudie, 8 

Pit.kerrald, 137, 161 

Pitma. glassy, 43 

Pitma.glaster, 137 

Pit.maiu, 269 

Pitmean, Paitmayne, Pet- 

mayue, 8, 208 
Pit.ourie, 208, 263 
Pitten. trail, 8, ix. 
Pitt.owrie, 137 
Pit.youlish, 137, 181, 31!) 
Place Names of Elginshire, 313 
Plare Names of Ross and 

Cromarty, 303 
Place Names of Scotland, 327 
.Pladaig, 56 
Pliny, 67 
.Plocktou, 61 
.Pluscarden, 127 
Poll-an-Tarie, 62 
Pol.maly, 153 
Poisons, 22 
.Porin, 174 
Port-a-Chullin, 61 
Port .Chaligaig, 17 
Poi't-Eorna, 61 
Port na.cloiclie, 61 
Port.rec, 34, 35, 156, xxv. 
Powrie, 174 
preas, 140 

Preas-.niucracli, 140, 276 
Priory of Kingussie, 203 
Proaig, 354 
.Pronsy, 18 

Pteroton Stratopedon , 315 
Ptolemv, 5, 6, 7, 32, 53, 67, 68. 

.Purin, 174 
Pyble, 105 

Qiieys, 107 
.Quidiuish, 105 
.Quiness, 107 
.Quinigh, 107 

.Quishftdcr, 107 
Qwbewy], Clan, 215 


.Raasay, 36, 75, 102, 169, xiv„ 

rabhan, 308 
.Raernisli, 102 
Raig.beg, 158 
Raig.more, 158 
.Raisaburg, 39 
Rait, 204, 205 
Raitts, 265, 266 
Ra.lia, 277 
.Ramasaig, 37, 171 
Ramsden, Sir John, 205 
.Ranigdale, 90 
Rankin, 29 
Raoiceadal. 355 
.Rarnish, 102 
rat, 141, 158 
.Rater, 24 
.Ratter, 24 
Rauueyjar, 75 
Reay, 10, 11, 25, x. 
-Redigill, 16 
Reef, 106 
Reilig, 162 
.Reisgill, 16, xi. 
.Reiveal, 38 
.Reraig, 57 
.Resort, 93, xx. 
Rhys, Professor, 27 
•RiabhaoTian, 62 
.RibigHl, 14 
Ricina, 67 
Rickin, 29 
Rif, 106 
Rigg, 180 
.Rimisdale, 15, xi. 
Riochnabegg, 262 
Risary, 173. 291 
.Rispond, 17 
Robert III., 15 
Robin, 29 
Robson, 22 
Rodel, 90, 177, xx 
.Roeness, 90 
.Rogart, 7, 10, ix. 
Ro.hallion, 176 
.Roineval, 38 
.Roishader, 171 



.Roua, 16, 36, 75 

Ro-^nierta, 6 

Roe.neath, 285 

Boss, 129, 307 

Ross & Cromarty, Place Names 

of, 303 
Ross, Earl of, 33, 34, 53 
.Rossal, 17, xiii. 
.Rossay, 169 
Rosses, 54 
.Rothes, 319 
Rothie-, 141 

Rothie.moon, 181, xxvii. 
Rothie. murchus, 158, 190, 193 
Roweardmale, 15 
Rowestoreuastynghe, 15 
Rudha ua h-Aoirinn, 350 
Rue. more, 61 
Ruigh-.Aiteachain, 280 
Ruigh an Roig, 251 
Ruigh-da-ros, 156 
Ruigh-liouiitaig, 280 
Ruigh na h-Eag, 250 
Rum, 76, 159, 349 
.Rusigarry, 96, xxi. 
.Ruthven, 277 
Ry-.voau, 320 


.Saasaig, 37 

.Saddel, 353 

Sail-riabhach, 62 

St. Cattail, 4 

Saint Columba, 5, 32, 42 

St. Drostan's Chapel, 264 

— Pindlagau, 345 

— Finnan, 46 

— Kilda, 73, 79, 177 

— Munn'.s Isle, 46 
.Sallachy, 60 
.Sanda, 15 
.Sandray, 75, 169 
.Saraig, 306 

.Sandwood, Loch, 18, 117 
.Sandsair, 107 
.Sandwick, 117 
.Sango-beg, 16 

• Sandaide, 17 
saothair, 308 

— -more, 16 
Sauchie-burn, 60 

.Scaftigill, 97 

.Scaloway, 113 

.6calpa, Scaipaidh, Soalpay, 

36, 57, 92, 169 
.Scauiport, 183, xxvii. 
.Scapa, 91 
.Scaravay, 75 
.fcscarba, 75 
.Scarista, 111 
Scarp, 73, 92 
Scathach, 33 
.Scavaig, R., 37, xv. 
Schie.hallion, 176 
.Sc-iberscross, 18 
Sco, 38 

Score, Sguvr, 38 
. Scorrybreck, 38 
Scotland Fjord, 342 
.Scourie, 18, xiii. 
.Scoval, 38 
.Scrabster, 12 
Scyuend, 10 
.Seadair, 171 
.Seaforth, L., 83 
Seaforth, Lord, 55 
.Seauchreag, 61 
Seil, 77, 354 
Seil-Phaill, 21, 22 
Seil-Thomas, 21, 22 
Seil-Wohan, 21 
.seipeal, 284 
•Selback, 32 
Seth, 225 
Setr, 17, 79, 106 
setter, 106 
Sgath, Skagi, 38 
Sgeir, 10, 38, 109 
Sgor Gaoithe, 261 
.Sguabach, 249 
-shader, 39, 171 
.Shadir, 106 
.Shageary, 172 
Shaw, 4 
.Shawbost, 83 
Shaws, 211 

— of Rothiemurchus. 206 
Shelibost, 84 
..Shellay, 74, 92 
.Shesgnau, 254 
.Sheshader, 108 
Shetland, 69 
.Shigra, 18 
.Shiuness, 16, 23, 



Shirra-beg, 273 

Shirra.more, 273 

.Siiisken, 254 

.Shiunaj 77 

.Shona, 7T 

.Shulishader, 39. 108 

.Siulista, 39 

.Shuna, 68, xviii. 

.Shurrery, 173, 290 

Sigurd of Orkney, 9, 13 

Siuciaii, 21 

Siol-Mliatliaiy, 23 

.sitheau, 309 

Sitbean Mor Dhail a' Caor- 

uiun, 252 
.Skaravat, 117 
.Skeabost, 171 
.Skegirsta, 110 
.Skelbo, 8, 13, 19 
.Skelbustar, 84 
Skene, Dr, 42 
Sker, 109 
Skericha, 18 
.Skernisb, 38 
.Skerra, Port, 18 
.Skerray, 18 
Skerri.-more, 109 
Sketis, Skitis, 68, 143 
.Skiary, 173, 291 
.Skibo, 11, 13, 21, 95 
Skinashiuk, L., 11, ix. 
.Skinid of Tongiie, 11 
.Skinuastadr, 11 
Skinner, 10 
.Skiport, 93 
.Skudiburg, 39 
.Skulamufi, 179 
-Skullomie, 18 
Skye, 32, 68. 71. 143, 145 
sleaghacli, 308 
Sleat, 34, 35, 155, xiv. 
Sletdale, 15 
.Sletell, 17 
Sliabh, An. 269 
Sliabli Loi-gach. 49 
Slochd, 321 
Smertae, 6 
.Smigel, 16, xii. 
.Smirisary. 173, 291 
Smiths, 211 
Smoo, 17 

.Sneachdach .Slinnean, 253 
.Snizort, 35, 37, 79, 93, 156 

Soa, 76, 79 
.Soarary, 173 
.Soay, 37, 168 
.Sodhulum, 170 
.Soillierie, 278 
Somerled, 70, 218 
.Sorby, 306 
.Soroba, 306, 354 
.S'Owerby, 306 
.Soyea, 15 

Spean, 43, 47, 143, 244 
Spey, 47, 143, 189, 238 
.Spiuuingdale, 14 
srath, 141 

Sron au loiair, 248 
Sron-na-Baruinn, 191 
-sta, 39 
Stac, 10 

Stack or Stakkv, 38 
Stadhr, 110 
Saffa, 77, 180 
.Staffin, 36, 180 
Stagiium Aporicuni, 151 
.Steallag, 250 
.St^^inisb, 101 
.Stelligarry, 172 
.Steunis, 101 
.Stimaravay, 117 
.Stockiuish, 103 
.Stockval, 38 
Stoer, 17 

Stokes, Dr. 7, 28. 69 
Stoneybridge, 178 
Stor-ass, 17 
Stornoway, 48, 116 
.Storniia. Locli. 116 
Storr. Ben, 38 
Strandabliat, 112 
Strath, 35, 155 
Strath. asgaig, 57, xvi. 
Strath. bogie, 202 
Strath. errick, 126. 163 
Strath. g-lass, 175 
.Strathie, 61, xvii. 
Strath.mashie, 242 
Strath. nairn. 11. 164 
Strath. naver, 3. 21, 25 
•'.-:^th p-efff>r, 129 
J'trath. sword ale. Strath.oidil. 

34. 156 
Strath. iiridale, 14 
.Strolanuis, 179 
Stroineferry, 56 



StTbnd, 112 
Strone, 204, 205, 369 
Suidhe, An, 251 
.Suileag, R., 48 
.Suilven, 15, 108 
.Suiagil, 16, xii. 
.Suisniah, 38 
Sule Skerry, 86 
.Sulishader, 171 
.Sulisker, 108 
.Suiiart, 166, 355 
Sutherland, Earl of, 3 
.Swaniboat, 84 
.Swordale, 14, 89, 109, ziv. 
.Swyuasetter, 84 


.Tabost, 80 

.Tahay, 74 

.Talisker, 109, 179 

Tarlogan, Tarlagau, 161 

Talyour, 24 

.Taransay, 74, 168, xx. 

.Tarbert, 175 

.Tarradale, 10, 167 

.Tarskavaig, 37 

Tarvediiin or Tarvedrum, Cape 

of, 6 
Teampull ua h-Annaid, 162 
Teangue, 113, xxii. 
Temple-House, 162 
Terra Ethica, 68 
Teutons, 70 
.Thamnabbaidh, L., 80 
.Tbealasbhaidh, L., 80 
Thomas, Captain, 13, 73. 80, 

86, 95, 96, 98, 103 
Thor, 12 
Tborfin, 9 
Thori, 12 
Thorir, 12 
Thoretein the Red, 9 
.Thuncairidh, Beinu, 355 
Thurso Bay, 6 
.Tiauavaig, 37 
Tigh-au-Teanipuill, 162 
Tillie Sow. 267 
Tilli.fonr, 174 
Tilli.foiirie, 174. 263 
Tilli.pourie, 125, 263 
.Timsgarry, 96 

.liiig\()ld, 82 

-liugwall, 114 

tiobar, 320 

Tir an.drish, 49, 68. 69, 71, 77, xviii. 

Tir-etigean, 354 

tobar, 320 

Tobar na h-Annaid, 162 

Tobar-naBi-buadh, 177 

toft, 113 

.Tolsta, 111 

Tol.vah, 230 

Tom Euuau, 161, 236 

Tomin, 29 

Tomna.hurich, 124, 184, 299. 

Tong, 113 
Tongue, 9, 10 
.Torboll, 12, 13 
Tore, An, 255 
Torfnes, 316 
.Torgabost, 170 
Tor.gulbin, 47 
Torran-dow, 21 
.Torrisdale, 12 
.Torroboll, 12, 13 
.Torvaig, 37 
Tor.vean, 124 
Tot or Tobhta, 39 
.Totaichean Aulaidh, 113 
.Totaig, 37 
Tot.score, 113 
Totta, 113 
Touug, 113 

.Tralagill, 16, 97, xii., xxi. 
Trantle, 12 
.Trasary. 173, 291 
Treig, L., 47 
.Treshinish, 77 
Trina.four, 174, 263 
.Ti-odday, 36, 76, xiv., xix. 
Troll, 112 
Trollness, 76 
.Trommie, 237, 239. 259 
.Trotternish, 34, 35, 76. 166 
Truim, 240 
.Trumsgarry, 97, 172 
Tulli.bardine. Morays of, 19 
Tullich, 213 
.TuUoch, 49 
Tiilloch.gorum, 193 
.Tulnen, 239 



tiiuga, 113 

Tresis, 143, 238, 322, 274 


[Jaimli Mhor, Au, 267 
.Uath lochau, Na t', 27] 
.Udalan, Glen, 61 
.Udlaman, Ben, 6] 
.Udrigill, 97 
-TTdrinfle, 179 
Uig, 118 
.Uignish, 38 
.Uigshaaer, 108, 171 
Uinneag Coire an Eich, etc. 

.Uisgeval, 171 
• Uisgneval. 171 
Uist, 67; 76 
Uiy, 91 
.Ulbster, 12 
.Ulhava (see Ulva), 56 
.Ulladale, 12 
.Ullapool, 12, 95 
.Ulliuish, 38 
.Ullipsdale, 12, 15 
.Ulva, 56, 77, xvi. 
.Unabol, 13 
.Unapool, 12 
.Unes, 16 
.Ungshader, 108 
Uni, 12 
.Unish, 38 
TJntliank, 317 
Uppat, 17 
Urie, 238 
.Urigil, 16 
-Urquhart, 185, .325 
.Urray, 185 


.Vaccasary, 173, 291 
Vacomagi, 143 
vagr, 113, 115 
Vail, .38 
.Vallay, 75 
.Valtos, 81 
Vanisary, 173, 291 
.Vanisli, 103 

Varar, 143 
.Varkasaig, 37 
Yarragill, 97 
.Vassipoll, 95 
•Vatersay, 75 
.Vatisker, 109, xxi. 
Veir, 24 
Vernacular Inscriptions of tlie 

Ancient Kingdom of Albnn, 

Verubium, 7 
Vestmann-hafn, 105 
.Vidigill, 38 
vik, 13, 17, 117 
.Vikisgill, 38, 97 
Vinegar Hill, 181, 258 
Virvedrura, Cape, 7 
Volcae, 72 
Vollr, 17 
.Volovig, 37 
Volsas, 53 
Vomanivs, 47 


• Waldbrek, 81 

.Walgarth, 81 

-wall, 114 

Walls, 114 

Wardlaw, 157 

. Water nish, .34, 35, 166 

.Wattin, 29 

Welsh, 72 

Western Isles. 68, 69 

Whitby, 10 

Whiterigs, 317 

Wiay, 36, 75. xiv. 

Wiciv, 117 

Widewall. 114 

Wilkin, 29 

William, King, 19 

Windhouse, 81 

Wolf of Badeuoch, 34. 197-8-9, 

Worgan, 26 
Wyvij?, 130 

Yarrow, 166 
York, 134, 300 

Bht>' Biichravi, 7s 6d. 

The School Gaelic Dictionary. 


8vo. Wrapper, Is Net. 

Drum-na-Keil (The Ridge of the Burial Place). 


Demy 8vo., buckram, 12s 6d ; % Calf, 25s, with map and portrait 
after the painting by Sir George Reid. 

The Highlanders of Scotland. 

Their viigiu, History, and Antiquities, with a Sketch of their Manners 
and Customs, and an acceunt of the Clans into which ihey were 
divided, and of the state of Societj which existed among them. 


The original text and notes intact, with additional notes, bringing the 
work, where necessary, to the position held in the same author's 
"Celtic Scotland,^' by ALEXANDER MacBAIN, M.A., LL.D., 
author of an Etymological Dictiouary d the Gaelic Language. 

Deiny 8vo., 10s 6d. 

Celtic Mythology and Religion. 

With Chapter upon DRUID CIRCLES and CELTIC BURIAL, with 
Introductory Chapter and Notes by Professor W. J. WATSuN. 


Dem,y Svo,, Wrapper, 2s 6d. 

Etymology of the Principal Gaelic National Hames, 
Personal Names and Surnames. 

To which IS added a DISQUI81TI0N on :: :: 



Demy 8vo., 21s ; '< Morocco, 32s 6d. 

Etymological Dictionary of the Gaelic Language. 


Post 8vo., 5s. 

Culloden IVloor and Story of the Battle. 

With Description of the STONE 

By the late PETER ANDERSON of Inverness. 

Post 8vo., 2s. 

The Massacre of Gkncoe and the Campbells of Glenlyon. 

and Professor BLACKIE. 

Demy Svo., cloth, gilt top, uncut, with map 
and illustrations, 3s 6d. 

Notes on the Lake of Menteith. 


Crown Svo,, 3s 6d. 

The Feuds of the Clans. 


Together with the History of the Feuds and Conflicts among the 
Clans in the Northern Parts of Scotland and in the Western Isles. 
From the year MXXXI. unto MDCXIX. 

Illustration: Coloured Frontispiece after Erskine Nicol, R.S.A. 

Crozvn Svo , 7^ 6d net. 

Scottish Life and Character 

In Anecdote and Story. 


Cloth, 3s 6d. 

The Prophecies of the Brahan Seer. 

With Introfluctory Chapter hy ANDREW LANG. 

With Illustrations of Smuggling Bothies, Distilling 
Utensils, etc. 8vo. Post, 3s. 

Smuggling in the Highlands. 

An Account of Highland Whisky with Smuggling Stories and Detections. 

Crown 8vo , 3s 6d. 

The Romance of Poaching in the Highlands ot Scotland. 

As Illustrated in the lives of JOHN FARQUHARSON and 
ALEXANDER DAVIDSON, the last of the Free Foresters. 


Crown 8vo., 3s 6d. 

The History of the Highland Clearances. 

With a New Introduction by IAN MACPHERSON, M.P. 

dvo. Wrapper, 2s; Postage, 2d. 

Songs of the Gael. 


Highland Second Sight ^/^ 


Prophecies of Coinneach Odhar and the Petty Seer. 


A New Issue, Demy Svo., 3s 6d. 

Highland Superstitions. 


With an Introductory Chapter upon Superstitions and their Origin 


DA 869 .IVI12 



Macbain, Alt 

Bxander , 


Place names 

, Highland 



islands of 

Scot land 


AZK-5953 (awih)